MomentumPro
Trends & Momentum Dashboard
This panel gathers global macroeconomic indicators — growth, inflation, rates, FX, and risk perception. Together, they form the backdrop that influences all asset classes.
Sentiment & News of the Week
Sentiment Index
Neutral
Negative
Neutral
Positive
10%
Pos
78%
Neu
12%
Neg
Source: VADER (sentiment), Perplexity AI (geopolitical), Claude API (commentary)
Global Macro Map
Interactive map with macroeconomic indicators by country — GDP, inflation, rates, FX, and bond yields[?]. Toggle between layers (type and period) and click any country to open the detail panel.
How to read this map: Each bubble is a country. Color indicates the selected indicator's value (green/warm = high, blue/cool = low). Size reflects relative GDP. Use the layer buttons to switch between indicators and periods. Click a country to see all data.
▸Methodology Note — Macro Map
What is this map?
A global view of macroeconomic indicators by country, updated weekly. Each bubble represents a country, colored and sized according to the selected layer (stock index, FX, inflation, etc.).
Data sources
• Stock indices — 38 countries via EODHD (1M, 3M, YTD, 12M returns)
• FX — ~40 pairs vs USD from forex.db (more precise than FRED)
• Macro — GDP, inflation, rates, yields, unemployment, debt/GDP via FRED (~45 countries)
• Geopolitical — AI-generated sentiment (Perplexity) per country
How to read?
Use the layer buttons (type and period) to switch between indicators. Click any country to open a detail panel with all available indicators. Warm colors = high values, cool colors = low values.
A global view of macroeconomic indicators by country, updated weekly. Each bubble represents a country, colored and sized according to the selected layer (stock index, FX, inflation, etc.).
Data sources
• Stock indices — 38 countries via EODHD (1M, 3M, YTD, 12M returns)
• FX — ~40 pairs vs USD from forex.db (more precise than FRED)
• Macro — GDP, inflation, rates, yields, unemployment, debt/GDP via FRED (~45 countries)
• Geopolitical — AI-generated sentiment (Perplexity) per country
How to read?
Use the layer buttons (type and period) to switch between indicators. Click any country to open a detail panel with all available indicators. Warm colors = high values, cool colors = low values.
Source: FRED, EODHD, forex.db, Perplexity AI
Taylor Rule Monitor
Deviation between the Taylor Rule[?] prescribed rate and the actual policy rate for 29 economies. Bars to the right (gold) indicate looser monetary policy than prescribed; to the left (blue), tighter.
▸ Country Details
Convergence Projection
Expected interest rate path based on regression Δratet+h = α + β · deviationt. Band = 95% confidence interval.
How to read this chart: Each bar is a country. Gold bars to the right = rates below prescribed (loose policy). Blue bars to the left = rates above prescribed (tight policy). Longer bars indicate greater misalignment between actual rates and what economic conditions suggest.
▸Methodology Note — Taylor Rule
What is the Taylor Rule?
A formula created by economist John Taylor (1993) that calculates what a country's interest rate should be based on two variables: how much inflation is above or below the target, and how much the economy is above or below its potential (the so-called output gap).
How does it work?
The formula is: i = r* + π + 0.5·(π − π*) + 0.5·gap
• r* — neutral real interest rate (when the economy is in equilibrium)
• π — current inflation (annual CPI)
• π* — central bank's inflation target
• gap — output gap: how much GDP is above (+) or below (−) potential
The gap is estimated using the Hodrick-Prescott filter (λ=1600) on quarterly real GDP since 1995. The HP trend represents potential output; the percentage deviation is the gap.
What does the deviation mean?
• Positive deviation (gold) — real rates are below prescribed → looser monetary policy than recommended
• Negative deviation (blue) — rates are above prescribed → tighter monetary policy
What is it for?
Identifying which countries have rates misaligned with economic conditions — which can anticipate changes in monetary policy or movements in FX and equities.
A formula created by economist John Taylor (1993) that calculates what a country's interest rate should be based on two variables: how much inflation is above or below the target, and how much the economy is above or below its potential (the so-called output gap).
How does it work?
The formula is: i = r* + π + 0.5·(π − π*) + 0.5·gap
• r* — neutral real interest rate (when the economy is in equilibrium)
• π — current inflation (annual CPI)
• π* — central bank's inflation target
• gap — output gap: how much GDP is above (+) or below (−) potential
The gap is estimated using the Hodrick-Prescott filter (λ=1600) on quarterly real GDP since 1995. The HP trend represents potential output; the percentage deviation is the gap.
What does the deviation mean?
• Positive deviation (gold) — real rates are below prescribed → looser monetary policy than recommended
• Negative deviation (blue) — rates are above prescribed → tighter monetary policy
What is it for?
Identifying which countries have rates misaligned with economic conditions — which can anticipate changes in monetary policy or movements in FX and equities.
Source: FRED (GDP, CPI), BCB SGS 432 (Selic)
Global FX
Performance of major currencies against the US dollar across multiple periods. Positive returns indicate currency appreciation vs USD. The scatter plot shows the correlation[?] between FX and equities by country.
▸ Currency Details
| Currency | Rate [?] | 1S [?] | 1M | 3M | YTD | 12M | Loading [?] |
|---|---|---|---|---|---|---|---|
| Argentine Peso | 1398.2500 | -2.2% | +1.2% | +5.2% | +3.7% | -30.3% | — |
| Nigerian Naira | 1385.1500 | -0.1% | -0.3% | +3.4% | +4.2% | +9.9% | — |
| Norwegian Krone | 9.7388 | -0.4% | -0.7% | +3.2% | +3.4% | +6.4% | -0.114 |
| Brazilian Real | 5.2526 | -0.2% | +0.5% | +3.1% | +4.1% | +7.2% | -0.565 |
| Colombian Peso | 3666.9700 | +0.9% | +2.8% | +2.7% | +2.0% | +11.6% | — |
| Australian Dollar | 0.6852 | -0.5% | -2.6% | +2.4% | +2.7% | +9.4% | +0.248 |
| Chinese Yuan | 6.9073 | +0.1% | -0.1% | +1.2% | +1.2% | +5.0% | -0.904 |
| Malaysian Ringgit | 4.0200 | -0.7% | -1.9% | +0.8% | +0.9% | +9.7% | — |
| Israeli Shekel | 3.1702 | -1.2% | -2.6% | +0.6% | +0.5% | +14.6% | -0.511 |
| Vietnamese Dong | 26345.0000 | +0.0% | -0.6% | -0.2% | -0.2% | -2.8% | — |
| Singapore Dollar | 1.2907 | -0.4% | -1.0% | -0.3% | -0.4% | +4.2% | — |
| New Zealand Dollar | 0.5719 | -1.5% | -3.9% | -0.7% | -1.2% | -0.5% | +0.194 |
| Swiss Franc | 0.7984 | -0.3% | -2.1% | -0.8% | -0.7% | +9.4% | +0.005 |
| Mexican Peso | 18.0925 | -0.8% | -2.2% | -1.0% | -0.6% | +10.5% | -0.151 |
| Russian Ruble | 81.3000 | +0.1% | -4.7% | -1.2% | -3.2% | +3.5% | -0.067 |
| Canadian Dollar | 1.3927 | -0.5% | -1.8% | -1.4% | -1.5% | +2.1% | -0.274 |
| Indonesian Rupiah | 16986.5000 | -0.5% | -0.8% | -1.6% | -1.9% | -2.6% | -0.843 |
| Japanese Yen | 159.8521 | -0.1% | -1.3% | -1.9% | -2.0% | -8.1% | +0.003 |
| British Pound | 1.3172 | -1.2% | -1.4% | -2.1% | -2.2% | +1.0% | +0.101 |
| Euro | 1.1456 | -0.7% | -2.1% | -2.2% | -2.5% | +5.0% | -0.068 |
| Taiwan Dollar | 32.1230 | -0.5% | -1.2% | -2.4% | -2.5% | +3.3% | — |
| Romanian Leu | 4.4419 | -0.9% | -2.0% | -2.5% | -2.6% | +3.5% | -0.047 |
| Hungarian Forint | 336.3300 | +0.1% | -0.9% | -2.9% | -2.8% | +8.9% | -0.017 |
| Chilean Peso | 932.6800 | -0.3% | -2.6% | -2.9% | -3.6% | +2.4% | — |
| Philippine Peso | 60.6470 | -0.7% | -3.8% | -3.1% | -3.0% | -6.2% | — |
| Turkish Lira | 44.4684 | -0.3% | -1.2% | -3.3% | -3.5% | -17.3% | -0.817 |
| Swedish Krona | 9.5407 | -1.1% | -3.0% | -3.5% | -3.5% | +3.7% | +0.003 |
| Czech Koruna | 21.3810 | -0.6% | -1.8% | -3.8% | -3.9% | +6.9% | -0.000 |
| South African Rand | 17.1341 | -0.1% | -3.6% | -3.9% | -3.8% | +9.2% | — |
| Peruvian Sol | 3.4925 | -0.3% | -2.1% | -3.9% | -3.9% | +4.9% | — |
| Polish Zloty | 3.7359 | -1.1% | -3.2% | -4.2% | -4.0% | +3.5% | +0.060 |
| Thai Baht | 32.8500 | +0.3% | -3.9% | -4.3% | -4.3% | +4.2% | — |
| Indian Rupee | 94.2950 | -0.0% | -2.4% | -4.8% | -4.8% | -10.3% | -0.762 |
| Korean Won | 1529.9800 | -1.5% | -3.3% | -6.1% | -6.0% | -4.4% | +0.194 |
| Egyptian Pound | 54.4000 | -3.6% | -10.6% | -14.2% | -14.1% | -7.6% | — |
Positive returns = currency appreciated vs USD. 90d sparkline shows cumulative % change.
Loading: FX→equity transmission coefficient estimated via PanelOLS with country fixed effects and Driscoll-Kraay standard errors. Negative values indicate currency depreciation is associated with local stock market decline.
▸Methodology Note — Global FX
What does this section show?
The performance of major world currencies against the US dollar (USD), grouped by region. Positive returns mean the currency appreciated against the dollar.
Data sources
Daily quotes for ~40 currency pairs via EODHD, stored in forex.db. Returns calculated for 1-week, 1-month, 3-month, YTD, and 12-month periods. Sparklines show cumulative change over the last 90 days.
FX × equity correlation
The scatter plot crosses FX return (3M) with local equity index return. Pearson correlation (ρ) shows the degree of association: values near +1 indicate that when the currency appreciates, the stock market tends to rise as well.
What is it for?
Mapping which currencies are strengthening or weakening, and how that relates to local equity markets.
The performance of major world currencies against the US dollar (USD), grouped by region. Positive returns mean the currency appreciated against the dollar.
Data sources
Daily quotes for ~40 currency pairs via EODHD, stored in forex.db. Returns calculated for 1-week, 1-month, 3-month, YTD, and 12-month periods. Sparklines show cumulative change over the last 90 days.
FX × equity correlation
The scatter plot crosses FX return (3M) with local equity index return. Pearson correlation (ρ) shows the degree of association: values near +1 indicate that when the currency appreciates, the stock market tends to rise as well.
What is it for?
Mapping which currencies are strengthening or weakening, and how that relates to local equity markets.
Source: EODHD forex.db
When Currencies Fall, What Happens to Stocks?
We measure the daily impact of currency depreciation on each country's stock market using panel regression[?]. The more negative the score, the greater the vulnerability of local stocks to currency shocks.
Does This Effect Change Over Time?
The FX-equity relationship is not fixed. During crises it intensifies (lines plunge), in calm periods it weakens. Each line shows how each country's sensitivity evolved over the past 2 years.
Scenario Simulator
Select a country and simulate shocks to see the estimated impact on local stocks.
Estimated stock market impact
0.00%
via FX: 0.00%
via global factors: 0.00%
Based on econometric model with historical data. Results are estimates, not predictions.
Technical model details
The simulator combines two components estimated via panel regression (PanelOLS) with country fixed effects:
requity = βi × rfx + Σ γk × Gk + αi + ε
- βi — Country-specific FX-equity beta (Model A). Estimated with Driscoll-Kraay standard errors (Bartlett kernel, bandwidth=5), robust to cross-sectional and serial correlation.
- γk — Global factor coefficients (Model B): S&P 500, VIX, oil, gold, DXY. Estimated with country-clustered standard errors.
- αi — Country fixed effect (absorbed), captures structural level differences.
Data: daily returns from 38 countries, 36,884 obs (2010-01-05 — 2026-03-31). FX variable: local currency depreciation vs USD (positive = weakening). Winsorized at 0.1%/99.9%.
Econometric model: daily panel with country fixed effects and standard errors robust to autocorrelation and cross-market dependence. · 2010-01-05 — 2026-03-31 · 38 countries · 36,884 obs
▸Methodology Note — FX → Equity Panel
What is this analysis?
It measures how much each country's stock market reacts when its currency weakens. The idea is simple: in many countries, when the currency weakens, foreign capital leaves and stocks fall together. But the intensity of this reaction varies widely across countries.
How does it work?
We use panel regression with entity and time fixed effects, analyzing daily returns from 38 countries. Controls isolate global factors (S&P 500, VIX, gold, oil) to measure the pure effect of currency on local stocks. Four robustness models confirm results:
• Base model (FX → equity)
• With global controls (SPY, VIX, GLD, CL)
• With structural interactions (real rates, export profile)
• Full model (all factors)
What amplifies the effect?
Countries with high real rates or heavy commodity export dependence tend to suffer more: speculative capital flees at the same time the currency weakens, amplifying the stock decline.
How to read?
Bars to the left (red) = stocks fall when currency weakens. Longer bars mean higher sensitivity. Stars (★) indicate statistical significance.
It measures how much each country's stock market reacts when its currency weakens. The idea is simple: in many countries, when the currency weakens, foreign capital leaves and stocks fall together. But the intensity of this reaction varies widely across countries.
How does it work?
We use panel regression with entity and time fixed effects, analyzing daily returns from 38 countries. Controls isolate global factors (S&P 500, VIX, gold, oil) to measure the pure effect of currency on local stocks. Four robustness models confirm results:
• Base model (FX → equity)
• With global controls (SPY, VIX, GLD, CL)
• With structural interactions (real rates, export profile)
• Full model (all factors)
What amplifies the effect?
Countries with high real rates or heavy commodity export dependence tend to suffer more: speculative capital flees at the same time the currency weakens, amplifying the stock decline.
How to read?
Bars to the left (red) = stocks fall when currency weakens. Longer bars mean higher sensitivity. Stars (★) indicate statistical significance.
Source: EODHD (indices, FX), FRED (global factors)
Risk Perception
How much stress is in the financial system right now? This composite index combines 20 volatility and credit indicators (VIX, commodity volatility, credit spreads, risk ETFs) into a unified view of systemic risk[?]. The chart shows which dimension (equities, credit, EM) is dominating stress.
68
High Fear
How to read this chart: The stacked area chart shows systemic risk evolution over time. Each colored band represents a risk category (equity volatility, credit, EM, etc.). When a band expands, that dimension is dominating stress. The radar on the right compares the current profile with 3 months ago.
▸Methodology Note — Systemic Risk Perception
What is systemic risk?
Risk that affects the financial system as a whole — not just a single asset or sector. When systemic risk rises, all risky assets tend to fall together.
How do we measure it?
We combine 20 series in a unified analysis via PCA (Principal Component Analysis):
• 8 CBOE volatility indices — VIX (US equities), OVX (oil), GVZ (gold), VXEEM (EM), VXFXI (China), VXEFA (developed ex-US), MOVE (bonds), TYVIX (treasuries)
• 12 credit proxy ETFs — HYG, JNK (high yield), LQD, VCIT (investment grade), KRE, KBE (banks), EMB, PCY (emerging), TLT, IEF (treasuries), SHY (short-term), BKLN (loans)
PCA extracts the "common factor" that explains most of the co-movement across these series — that factor is our systemic risk index.
How to read the chart?
The stacked area chart decomposes each category's contribution (equity volatility, commodities, credit, emerging markets, etc.) to total risk. When a slice expands, that dimension is dominating market stress.
Risk that affects the financial system as a whole — not just a single asset or sector. When systemic risk rises, all risky assets tend to fall together.
How do we measure it?
We combine 20 series in a unified analysis via PCA (Principal Component Analysis):
• 8 CBOE volatility indices — VIX (US equities), OVX (oil), GVZ (gold), VXEEM (EM), VXFXI (China), VXEFA (developed ex-US), MOVE (bonds), TYVIX (treasuries)
• 12 credit proxy ETFs — HYG, JNK (high yield), LQD, VCIT (investment grade), KRE, KBE (banks), EMB, PCY (emerging), TLT, IEF (treasuries), SHY (short-term), BKLN (loans)
PCA extracts the "common factor" that explains most of the co-movement across these series — that factor is our systemic risk index.
How to read the chart?
The stacked area chart decomposes each category's contribution (equity volatility, commodities, credit, emerging markets, etc.) to total risk. When a slice expands, that dimension is dominating market stress.
Source: FRED (VIX, VXN, VXEEM, GVZ, OVX), EODHD (credit ETFs)
Weekly Reading
In the week of 02/04/2026, momentum highlights stocks like CF (88), PBR (84), CE (83), and PETR3.SA (82), as well as ETFs like FNGD.US (85) and TSLQ.US (83), with dominant sectors Basic Materials and Energy, both with 3 representatives at the top. Institutional money flows into entries in SPDR S&P 500 ETF Trust (+10290M), Invesco QQQ Trust (+7802M), and SPDR Gold Shares (+2065M), while recording outflows from iShares Core S&P 500 ETF (-3028M) and United States Oil Fund LP (-2684M). The market regime is risk-on, with a 92% probability, reinforcing risk appetite.
This panel provides insight into which types of assets are performing better or worse — and why. All analyses are based on robust quantitative methodologies widely used in academic and institutional settings.
Assets in Momentum (Uptrend)
The system uses fuzzy logic[?] to evaluate each asset: instead of rigid rules (e.g., "above 20-day moving average → bullish"), it assigns membership degrees to various bullish indicators. 11 rules combine these degrees to generate a signal (strong or moderate) with a confidence between 0 and 1. A momentum score is generated by weighting each evaluated indicator. The 6 assets with the highest score are shown in each group below. Click "View Details" to see the recent price chart. Below, we present a backtest of the methodology to assess whether the score predicted positive returns retrospectively.
▸ Top 8 International Stocks — Fuzzy Momentum
CF
Basic Materials · Agricultural Inputs
MODERATE SIGNAL
▸
CF Industries Holdings, Inc., juntamente com suas subsidiárias, atua na produção de amônia na América do Norte, Europa e internacionalmente. Ela opera por meio dos segmentos Ammonia, Granular Urea, UAN, AN e Other.
Perf 1M
+31.9%
Perf 6M
+52.0%
Sharpe 1Y
1.90
P/E11.7
Fwd P/E12.3
Margem20.5%
ROE23.4%
Div Yield1.88%
D/E0.82
Mkt Cap$16.3B
PBR
Energy · Oil & Gas Integrated
MODERATE SIGNAL
▸
Petróleo Brasileiro S.A. - Petrobras explora, produz e vende petróleo e gás no Brasil e internacionalmente.
Perf 1M
+20.2%
Perf 6M
+61.9%
Sharpe 1Y
1.21
P/E6.0
Fwd P/E6.4
Margem15.8%
ROE19.0%
Div Yield25.92%
D/E0.89
Mkt Cap$84.3B
CE
Basic Materials · Chemicals
MODERATE SIGNAL
▸
A Celanese Corporation produz e vende polímeros de engenharia em todo o mundo. Opera através dos segmentos Engineered Materials e Acetyl Chain.
Perf 1M
+29.2%
Perf 6M
+57.3%
Sharpe 1Y
0.13
Fwd P/E10.2
Margem-12.2%
ROE-22.5%
Div Yield0.23%
D/E0.36
Mkt Cap$5.6B
NGD Stale (2026-03-23)
Basic Materials · Gold
STRONG SIGNAL
▸
New Gold Inc., uma empresa intermediária de mineração de ouro, se dedica ao desenvolvimento e operação de propriedades minerais no Canadá. Ela explora principalmente depósitos de ouro, prata e cobre.
Perf 1M
+0.3%
Perf 6M
+77.3%
Sharpe 1Y
3.63
P/E32.5
Fwd P/E9.0
Margem20.1%
ROE22.3%
D/E0.32
Mkt Cap$8.3B
TDW
Energy · Oil & Gas Equipment & Services
STRONG SIGNAL
▸
Tidewater Inc., juntamente com suas subsidiárias, fornece embarcações de apoio offshore e serviços de suporte marítimo à indústria de energia offshore por meio da operação de uma frota de embarcações de serviço marítimo offshore em todo o mundo. A empresa oferece suporte nas fases de exploração d...
Perf 1M
+2.5%
Perf 6M
+51.8%
Sharpe 1Y
1.42
P/E28.2
Fwd P/E36.4
Margem24.7%
ROE27.0%
D/E0.48
Mkt Cap$4.2B
BP
Energy · Oil & Gas Integrated
MODERATE SIGNAL
▸
BP p.l.c., uma empresa de energia integrada, atua no negócio de petróleo e gás em todo o mundo. A empresa opera por meio dos segmentos Gas & Low Carbon Energy, Oil Production & Operations e Customers & Products.
Perf 1M
+20.0%
Perf 6M
+36.3%
Sharpe 1Y
1.14
P/E1942.0
Fwd P/E14.4
Margem0.0%
ROE1.7%
Div Yield5.14%
D/E1.59
Mkt Cap$99.5B
OXY
Energy · Oil & Gas E&P
MODERATE SIGNAL
▸
A Occidental Petroleum Corporation, juntamente com suas subsidiárias, atua na aquisição, exploração e desenvolvimento de propriedades de petróleo e gás nos Estados Unidos e internacionalmente. Opera através dos segmentos de Petróleo e Gás e Midstream e Marketing.
Perf 1M
+22.2%
Perf 6M
+37.7%
Sharpe 1Y
0.77
P/E39.7
Fwd P/E26.7
Margem10.8%
ROE5.9%
Div Yield1.79%
D/E0.66
Mkt Cap$52.9B
CHRD
Energy · Oil & Gas E&P
MODERATE SIGNAL
▸
A Chord Energy Corporation opera como uma empresa independente de exploração e produção nos Estados Unidos. A empresa se dedica à aquisição, exploração, desenvolvimento e produção de petróleo bruto, gás natural e líquidos de gás natural na Bacia de Williston.
Perf 1M
+29.1%
Perf 6M
+44.5%
Sharpe 1Y
0.59
P/E38.6
Fwd P/E0.0
Margem1.0%
ROE0.5%
Div Yield4.63%
D/E0.19
Mkt Cap$6.5B
▸ Top 4 Brazilian Stocks — Fuzzy Momentum
PETR3.SA
Energy · Oil & Gas Integrated
MODERATE SIGNAL
▸
Petróleo Brasileiro S.A. - Petrobras explora, produz e vende petróleo e gás no Brasil e internacionalmente.
Perf 1M
+27.9%
Perf 6M
+52.5%
Sharpe 1Y
1.08
P/E7.9
Fwd P/E6.5
Margem15.8%
ROE19.0%
Div Yield7.58%
D/E0.89
Mkt Cap$567.9B
JALL3.SA
Consumer Defensive · Confectioners
STRONG SIGNAL
▸
Jalles Machado S/A produz, comercializa e exporta açúcar, etanol e outros subprodutos de cana-de-açúcar. A empresa oferece açúcar cristal e orgânico; etanol hidratado, anidro, industrial e orgânico; produtos sanitizantes; e levedura.
Perf 1M
+20.1%
Perf 6M
+28.5%
Sharpe 1Y
-0.41
Fwd P/E5.4
Margem-3.2%
ROE-4.0%
Div Yield1.71%
D/E3.21
Mkt Cap$914M
BNBR3.SA
Financial Services · Banks - Regional
STRONG SIGNAL
▸
Banco do Nordeste do Brasil S.A. opera como um banco de desenvolvimento regional na América Latina.
Perf 1M
+1.2%
Perf 6M
+28.1%
Sharpe 1Y
0.69
P/E4.3
Fwd P/E0.0
Margem30.0%
ROE19.4%
Div Yield2.66%
D/E1.87
Mkt Cap$12.0B
SEER3.SA
Consumer Defensive · Education & Training Services
MODERATE SIGNAL
▸
Ser Educacional S.A., juntamente com suas subsidiárias, opera cursos de graduação, pós-graduação e formação profissional presenciais e a distância, além de outras áreas relacionadas à educação no Brasil. Oferece cursos de graduação, pós-graduação, presenciais e de aprendizagem digital, e cursos p...
Perf 1M
+0.2%
Perf 6M
+13.1%
Sharpe 1Y
2.61
P/E13.7
Fwd P/E9.6
Margem5.1%
ROE8.6%
Div Yield1.33%
D/E1.16
Mkt Cap$1.5B
▸ Top 8 ETFs — Fuzzy Momentum
FNGD.US
ETF · ETF
MODERATE SIGNAL
▸
O ETF FNGD.US (MicroSectors FANG+ Index -3X Inverse Leveraged ETN) é um ETN emitido pela BMO que busca replicar o retorno diário inverso e alavancado em -3x do NYSE FANG+ Index, composto por 10 ações igualmente ponderadas de empresas de tecnologia e consumo discricionário de alto crescimento, com...
Perf 1M
+28.0%
Perf 6M
+1588.4%
Sharpe 1Y
0.80
TSLQ.US
ETF · ETF
MODERATE SIGNAL
▸
O AXS TSLA Bear Daily ETF (TSLQ) é um fundo negociado em bolsa que fornece exposição inversa (-2x) aos movimentos diários das ações da Tesla, buscando lucrar quando o preço das ações da Tesla cai. O fundo investe em mercados de ações públicas através de derivativos, operando no setor de veículos ...
Perf 1M
+25.2%
Perf 6M
+234.5%
Sharpe 1Y
0.81
UVXY.US
ETF · ETF
MODERATE SIGNAL
▸
O ETF UVXY (ProShares Ultra VIX Short-Term Futures ETF) investe em contratos futuros de curto prazo do VIX, o índice de volatilidade do S&P 500, com exposição alavancada de 1,5x ao S&P 500 VIX Short-Term Futures Index, que replica posições nos próximos dois contratos futuros mensais com média pon...
Perf 1M
+53.6%
Perf 6M
+481.2%
Sharpe 1Y
0.81
SQQQ.US
ETF · ETF
MODERATE SIGNAL
▸
O ETF SQQQ.US (ProShares UltraPro Short QQQ) é um fundo negociado em bolsa alavancado e inverso que investe em derivativos, como swaps e futuros, para buscar resultados diários equivalentes a -3x o desempenho do Nasdaq-100 Index, um índice de 100 das maiores empresas não financeiras listadas na N...
Perf 1M
+26.2%
Perf 6M
+482.9%
Sharpe 1Y
1.11
TSLZ.US
ETF · ETF
MODERATE SIGNAL
▸
O ETF TSLZ.US (T-Rex 2X Inverse Tesla Daily Target ETF) investe em derivativos que buscam replicar o inverso duplo (-2x) do desempenho diário das ações da Tesla Inc. (TSLA), listada na NASDAQ. Ele foca no setor de fabricantes de automóveis nos Estados Unidos, dentro da categoria de consumo discri...
Perf 1M
+24.7%
Perf 6M
+2313.8%
Sharpe 1Y
0.76
SRTY.US
ETF · ETF
MODERATE SIGNAL
▸
O ETF SRTY.US (ProShares UltraPro Short Russell2000) é um fundo de índice inverso alavancado que busca entregar -3x o desempenho diário do índice Russell 2000, composto por cerca de 2.000 pequenas empresas americanas de capitalização de mercado abaixo de US$ 12 bilhões, selecionadas do Russell 30...
Perf 1M
+26.3%
Perf 6M
+275.1%
Sharpe 1Y
0.74
TECS.US
ETF · ETF
MODERATE SIGNAL
▸
O ETF TECS.US (Direxion Daily Technology Bear 3X Shares) é um fundo negociado em bolsa que busca entregar -300% do desempenho diário do Technology Select Sector Index, utilizando derivativos como swaps e futuros para uma estratégia de venda a descoberto alavancada. Ele investe indiretamente em aç...
Perf 1M
+25.6%
Perf 6M
+19.8%
Sharpe 1Y
0.80
TSLS.US
ETF · ETF
MODERATE SIGNAL
▸
O ETF TSLS.US (Direxion Daily TSLA Bear 1X Shares) é um fundo inverso que busca resultados diários equivalentes a 100% do inverso (-1x) do desempenho das ações comuns da Tesla, Inc. (TSLA), utilizando derivativos como swaps para criar exposição short no setor de veículos elétricos e componentes d...
Perf 1M
+1020.6%
Perf 6M
+1072.5%
Sharpe 1Y
0.76
Search Fuzzy Score
Search any stock or ETF in the universe to see its composite score, percentile, and position in the distribution.
Walk-Forward Backtest
To test whether the system really works, we went back in time: each Friday over the last 52 weeks, we recalculated scores using only data available on that date (no peeking into the future). The top 6 stocks + 6 ETFs were selected and then we measured what actually happened with those assets in the following 1, 2, and 3 months. The 3 indicators below summarize the 3-month result: the average return of the picks, how much they beat the S&P 500, and in how many weeks the picks beat the index (Win Rate — above 50% means the system got it right most weeks).
Average Return 3M
+10.6%
Excess vs S&P 500
+5.3%
Win Rate vs S&P 500
+66.7%
% of weeks that beat SPY
View Backtest Details
Source: EODHD (historical prices)
▸Methodological Note — Fuzzy Logic Momentum
What is fuzzy logic?
In traditional logic, a statement can only be true or false. In fuzzy logic, things can be partially true. For example: an asset priced 2% above its 20-day moving average is not "completely above" or "completely below" — it has an intermediate degree of membership in the "above average" group. This allows the system to capture nuances that binary rules would miss.
How does the momentum score work?
The system evaluates 5 indicators for each asset, each receiving a degree between 0 and 1:
• Weekly momentum — is the asset rising, falling, or flat?
• Position vs. 20-day moving average — is the price above or below the recent trend?
• Relative volume (10 days) — is trading volume above normal? (more people buying/selling)
• RSI (14 days) — is the asset overbought, oversold, or in a neutral zone?
• Trend type — is the trend consistently bullish, a reversal, or undefined?
11 rules combine these degrees to generate a buy signal (strong or moderate) with a confidence level. The final score weights each indicator according to its predictive importance.
Illustrative example:
Imagine an asset with the following readings: weekly momentum = 0.85 (strong rise), position vs. MA20 = 0.70 (well above average), relative volume = 0.60 (above normal), RSI = 0.55 (neutral-high zone), trend type = 0.90 (consistent uptrend). The 11 rules evaluate these combinations — for example, "if momentum is high AND position vs. MA20 is high, then the signal is strong with high confidence". The weighted final score would be approximately: 0.85×35% + 0.70×25% + 0.60×15% + 0.55×10% + 0.90×5% + confidence×10% ≈ 0.74. This score is compared against all other assets to form the ranking.
In traditional logic, a statement can only be true or false. In fuzzy logic, things can be partially true. For example: an asset priced 2% above its 20-day moving average is not "completely above" or "completely below" — it has an intermediate degree of membership in the "above average" group. This allows the system to capture nuances that binary rules would miss.
How does the momentum score work?
The system evaluates 5 indicators for each asset, each receiving a degree between 0 and 1:
• Weekly momentum — is the asset rising, falling, or flat?
• Position vs. 20-day moving average — is the price above or below the recent trend?
• Relative volume (10 days) — is trading volume above normal? (more people buying/selling)
• RSI (14 days) — is the asset overbought, oversold, or in a neutral zone?
• Trend type — is the trend consistently bullish, a reversal, or undefined?
11 rules combine these degrees to generate a buy signal (strong or moderate) with a confidence level. The final score weights each indicator according to its predictive importance.
Illustrative example:
Imagine an asset with the following readings: weekly momentum = 0.85 (strong rise), position vs. MA20 = 0.70 (well above average), relative volume = 0.60 (above normal), RSI = 0.55 (neutral-high zone), trend type = 0.90 (consistent uptrend). The 11 rules evaluate these combinations — for example, "if momentum is high AND position vs. MA20 is high, then the signal is strong with high confidence". The weighted final score would be approximately: 0.85×35% + 0.70×25% + 0.60×15% + 0.55×10% + 0.90×5% + confidence×10% ≈ 0.74. This score is compared against all other assets to form the ranking.
Source: EODHD (prices, fundamentals, 4K symbols)
ETFs — Largest Investment Inflows and Outflows (Last Week)
Shows the ETFs that received the most and lost the most capital in the last week, measured by the change in average daily trading volume. Useful for identifying where institutional money is flowing.
▸ ▲ Top Inflows (7 days)
| ETF | Flow 7d | Change | Vol/day |
|---|---|---|---|
| +$10289.8M | +17.0% | $70861.5M | |
| +$7801.9M | +21.6% | $43936.0M | |
| +$2065.0M | +33.5% | $8238.3M | |
| +$1788.4M | +22.7% | $9670.5M | |
| +$1006.3M | +7.9% | $13706.5M |
▸ ▼ Top Outflows (7 days)
| ETF | Flow 7d | Change | Vol/day |
|---|---|---|---|
| $3027.5M | -26.4% | $8445.9M | |
| $2684.5M | -31.7% | $5777.4M | |
| $1481.4M | -22.0% | $5264.1M | |
| $1384.7M | -66.4% | $699.7M | |
| $1289.0M | -87.6% | $181.7M |
Source: EODHD (ETF prices, AUM, holdings)
Fund Performance and Alpha
Evaluates investment funds using the academic Fama-French model. The goal is to answer: does this fund truly generate value, or does it just ride known risks? Alpha (α)[?] measures the annualized return not explained by the model's 5 risk factors.
▸ Top 10 Funds — Global
| # | Ticker | Name | Alpha (α) | 1M | 6M | 12M | β Mkt | β SMB | β HML | β RMW | β CMA | R² |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ▸1 | IGR | +0.2762 | +3.8% | -2.3% | +14.8% | +0.8822 | -0.1789 | +0.4062 | +0.1949 | +0.0762 | +0.3640 | |
| ▸2 | NAD | +0.0552 | -1.0% | -5.5% | +5.4% | +0.2685 | +0.0368 | +0.0576 | +0.1724 | +0.0134 | +0.1896 | |
| ▸3 | JPC | +0.0505 | +2.4% | +3.3% | +20.1% | +0.5175 | -0.2498 | +0.0324 | +0.0624 | +0.0449 | +0.3767 | |
| ▸4 | NEA | +0.0232 | -1.4% | -5.5% | +5.3% | +0.2933 | +0.0609 | +0.0627 | +0.2530 | -0.0618 | +0.2145 | |
| ▸5 | DNP | +0.0156 | +1.5% | +5.9% | +22.4% | +0.6101 | -0.2138 | +0.1409 | +0.2309 | -0.0648 | +0.3697 | |
| ▸6 | CLM | -0.0896 | +15.5% | -5.6% | +23.7% | +0.7964 | -0.2314 | -0.2471 | +0.1300 | -0.1158 | +0.3991 | |
| ▸7 | USA | -0.1043 | +4.7% | -3.5% | +10.7% | +0.8144 | -0.3147 | -0.1398 | +0.0091 | -0.0546 | +0.6898 | |
| ▸8 | PTY | -0.1053 | +0.7% | -2.0% | +5.4% | +0.5480 | -0.2457 | +0.0410 | +0.2645 | +0.0889 | +0.4157 | |
| ▸9 | JQC | -0.1101 | +2.0% | -3.1% | +6.3% | +0.5199 | -0.2138 | -0.0314 | +0.1316 | -0.0191 | +0.3439 | |
| ▸10 | JFR | -0.1140 | +1.9% | -3.6% | +7.3% | +0.4780 | -0.2795 | +0.0330 | +0.0541 | -0.0523 | +0.3500 |
▸ Top 10 Funds — Brazil
| # | Ticker | Name | Alpha (α) | 1M | 6M | 12M | β Mkt | β SMB | β HML | β RMW | β CMA | R² |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ▸1 | VRTA11.SA | +0.0796 | +0.9% | +9.1% | +4.2% | +0.4420 | +0.3086 | -0.0559 | -0.1848 | -0.1708 | +0.1107 | |
| ▸2 | HGLG11.SA | +0.0388 | -1.4% | +10.2% | +5.8% | +0.3000 | +0.2255 | -0.0878 | +0.0196 | -0.0246 | +0.0865 | |
| ▸3 | NDIV11.SA | +0.0222 | -0.6% | +7.5% | +11.0% | +0.4903 | -0.8169 | +0.1395 | -0.0068 | +0.4087 | +0.5670 | |
| ▸4 | MXRF11.SA | +0.0209 | +2.1% | +8.0% | +3.8% | +0.2760 | +0.1356 | +0.0888 | -0.0209 | -0.0883 | +0.0682 | |
| ▸5 | CRAA11.SA | +0.0190 | +2.4% | +10.2% | +4.3% | +0.3261 | +0.0011 | +0.0232 | -0.2097 | -0.1930 | +0.0488 | |
| ▸6 | NSDV11.SA | +0.0068 | 0.0% | +6.8% | +10.2% | +0.5238 | -0.8206 | +0.1815 | +0.0519 | +0.3783 | +0.6449 | |
| ▸7 | CXAG11.SA | -0.0427 | +3.8% | +6.8% | +8.4% | +0.1879 | +0.2698 | -0.0612 | +0.0681 | -0.0644 | +0.0518 | |
| ▸8 | HGBS11.SA | -0.1132 | +0.8% | +8.5% | -0.6% | +0.4160 | +0.2524 | -0.2095 | -0.1035 | -0.1323 | +0.1142 | |
| ▸9 | APTO11.SA | -0.1235 | +1.7% | +1.9% | +2.1% | +0.2010 | +0.6560 | +0.0523 | +0.5617 | -0.0422 | +0.0705 | |
| ▸10 | XPML11.SA | -0.1565 | -1.3% | +4.3% | -8.7% | +0.4202 | +0.2378 | -0.0624 | +0.0291 | -0.0310 | +0.1002 |
Search Alpha for Any Asset
Search any stock, ETF, or fund in the ~4,000-asset universe to see its alpha, Fama-French risk factor exposure, and position in the distribution.
▸Methodological Note — Fama-French 5-Factor Model
The origin of the model
In 1992, economists Eugene Fama and Kenneth French published a study that revolutionized how we evaluate investments. They discovered that a stock's return depends not only on "did the market go up or down", but on other predictable patterns. Eugene Fama received the Nobel Prize in Economics in 2013 for this and other work on financial markets.
The core idea — in plain language
Imagine you want to evaluate whether a fund manager is truly skilled. If their fund returned 15% this year, it sounds great — but what if the entire market rose 20%? In that case, the manager actually underperformed the market. The Fama-French model goes further: it checks whether the fund's return can be explained not just by the market, but by 5 patterns (called factors) that historically generate returns:
• Market — the extra return for investing in stocks instead of risk-free bonds (how much the market as a whole went up or down)
• Size (SMB) — smaller companies tend to outperform giant ones over time, because they are riskier
• Value (HML) — "cheap" stocks (low price relative to company assets) tend to beat "expensive" ones (trendy companies with inflated prices)
• Profitability (RMW) — more profitable companies tend to deliver better returns
• Investment (CMA) — companies that invest conservatively (without overspending on expansion) tend to outperform
What is Alpha (α)?
After discounting these 5 effects, what remains is Alpha. If a fund has positive alpha, it means it generates returns beyond what would be expected given the risks it takes. This suggests genuine manager skill. If alpha is negative, the fund is destroying value — likely due to high fees, bad decisions, or poor timing.
How to read the table
• Alpha (α) — the annualized extra return (positive = good, negative = bad)
• β Mkt, β SMB, β HML, β RMW, β CMA — the fund's exposure to each factor (higher values = more exposed to that type of risk)
• R² — how much of the fund's behavior is explained by the model (0% = nothing, 100% = fully). Low R² may mean the fund has a very different strategy from the traditional stock market
In 1992, economists Eugene Fama and Kenneth French published a study that revolutionized how we evaluate investments. They discovered that a stock's return depends not only on "did the market go up or down", but on other predictable patterns. Eugene Fama received the Nobel Prize in Economics in 2013 for this and other work on financial markets.
The core idea — in plain language
Imagine you want to evaluate whether a fund manager is truly skilled. If their fund returned 15% this year, it sounds great — but what if the entire market rose 20%? In that case, the manager actually underperformed the market. The Fama-French model goes further: it checks whether the fund's return can be explained not just by the market, but by 5 patterns (called factors) that historically generate returns:
• Market — the extra return for investing in stocks instead of risk-free bonds (how much the market as a whole went up or down)
• Size (SMB) — smaller companies tend to outperform giant ones over time, because they are riskier
• Value (HML) — "cheap" stocks (low price relative to company assets) tend to beat "expensive" ones (trendy companies with inflated prices)
• Profitability (RMW) — more profitable companies tend to deliver better returns
• Investment (CMA) — companies that invest conservatively (without overspending on expansion) tend to outperform
What is Alpha (α)?
After discounting these 5 effects, what remains is Alpha. If a fund has positive alpha, it means it generates returns beyond what would be expected given the risks it takes. This suggests genuine manager skill. If alpha is negative, the fund is destroying value — likely due to high fees, bad decisions, or poor timing.
How to read the table
• Alpha (α) — the annualized extra return (positive = good, negative = bad)
• β Mkt, β SMB, β HML, β RMW, β CMA — the fund's exposure to each factor (higher values = more exposed to that type of risk)
• R² — how much of the fund's behavior is explained by the model (0% = nothing, 100% = fully). Low R² may mean the fund has a very different strategy from the traditional stock market
Source: EODHD (4K stocks), Fama-French 5-factor model
Thematic Portfolios
The universe's assets are grouped into thematic portfolios (momentum, diversified, defensive, dollar, gold, oil, etc.) based on how they behave together. Assets that rise and fall in similar patterns are placed in the same group. Select a portfolio from the menu to see its constituent assets. Click any point in the network to see asset details and its most related peers — if the asset belongs to another portfolio, the view switches automatically.
How to read this chart: each bar represents the portfolio's exposure to a Fama-French risk factor. Bars to the right (positive) indicate the portfolio benefits from that factor. Bars to the left (negative) indicate opposite exposure. For example, a high "Market" value means the portfolio tends to rise when the market rises; a negative "Size" value indicates a preference for large companies over small ones.
▸Methodological Note — Thematic Portfolios and Correlation Network
What is a correlation network?
Imagine each asset (stock or ETF) as a dot. When two assets tend to rise and fall together, we draw a line between them. The more similar their behavior, the thicker the line. The result is a visual map where similarly-behaving assets are close together, and assets with different behavior are far apart.
How are portfolios formed?
From this network, the system automatically identifies natural clusters — groups of assets that move in similar ways. Each cluster receives a thematic name describing the dominant behavior of its members: "momentum" (assets in uptrend), "defensive" (more stable assets), "dollar" (assets sensitive to exchange rates), etc.
What is it for?
This map helps you understand the real diversification of a portfolio. If all your assets are in the same group, they will likely fall together during stress. Assets from different groups tend to offset each other, reducing overall risk.
How to read the chart:
• Dots = individual assets (stocks or ETFs)
• Lines = correlation between two assets (thicker = more correlated)
• Colors = each color represents a different thematic portfolio
• Proximity = nearby assets behave similarly
• Distance = distant assets offer diversification from each other
Imagine each asset (stock or ETF) as a dot. When two assets tend to rise and fall together, we draw a line between them. The more similar their behavior, the thicker the line. The result is a visual map where similarly-behaving assets are close together, and assets with different behavior are far apart.
How are portfolios formed?
From this network, the system automatically identifies natural clusters — groups of assets that move in similar ways. Each cluster receives a thematic name describing the dominant behavior of its members: "momentum" (assets in uptrend), "defensive" (more stable assets), "dollar" (assets sensitive to exchange rates), etc.
What is it for?
This map helps you understand the real diversification of a portfolio. If all your assets are in the same group, they will likely fall together during stress. Assets from different groups tend to offset each other, reducing overall risk.
How to read the chart:
• Dots = individual assets (stocks or ETFs)
• Lines = correlation between two assets (thicker = more correlated)
• Colors = each color represents a different thematic portfolio
• Proximity = nearby assets behave similarly
• Distance = distant assets offer diversification from each other
Source: EODHD (returns, correlations), Fama-French 5-factor
REITs — Real Estate Investment Trusts
REIT market overview: performance by sub-sector, geographic comparison, and recent top performers.
▸ Performance by Sub-Sector
🇺🇸 United States
| Sector | Ret 1M | Ret 6M | Yield |
|---|---|---|---|
| Mortgage (20) | -6.2% | -7.0% | 13.2% |
| Hotel & Motel (10) | -6.4% | -4.0% | 4.8% |
| Specialty (11) | -7.4% | -1.5% | 4.2% |
| Retail (17) | -7.6% | +0.5% | 3.7% |
| Diversified (5) | -7.8% | -7.0% | 5.7% |
| Residential (12) | -8.4% | -19.1% | 6.1% |
| Healthcare Facilities (10) | -8.4% | +8.1% | 4.0% |
| Industrial (11) | -9.5% | +37.4% | 4.5% |
| Office (13) | -10.5% | -19.7% | 7.0% |
🇧🇷 Brazil
| Sector | Ret 1M | Ret 6M | Yield |
|---|---|---|---|
| Office (2) | +0.6% | +43.1% | 0.0% |
| Diversified (32) | -0.3% | +15.3% | 0.6% |
| Residential (2) | -0.8% | +1.8% | 0.0% |
| Industrial (2) | -1.5% | +10.3% | 0.0% |
| Specialty (4) | -1.5% | +0.2% | 0.0% |
| Retail (3) | -4.5% | -16.8% | 0.0% |
Source: EODHD (fundamentals_enrichment — REITs)
Market Regime
Identifies the current market state by analyzing 11 asset classes weekly: equities (SPY), value vs. growth (IWD−IWF), momentum (MTUM), quality (QUAL), long-term bonds (TLT), investment-grade credit (LQD), high-yield credit (HYG), emerging markets (EEM), volatility (VIXY), commodities (DBC), and gold (GLD). The model automatically detects the market's current "mood" — whether it is optimistic and accepting risk, cautious, or in protective mode.
Where We Are Now
Current reading of the 4 principal components
PC1
Risk Appetite
Risk-On 66%
Elevated risk appetite — equities, momentum and credit rallying together. Favorable environment for risk assets.
PC2
Duration / Rates
Easing 82%
Yields falling, bonds rallying — monetary easing or flight-to-quality environment.
PC3
Cyclical Rotation
Cyclical 70%
Cyclical rotation — value, commodities and inflation assets leading. Reflation trade.
PC4
Tail Risk
Alert 95%
Tail risk alert — elevated VIX and gold demand as refuge. Protection recommended.
How regime identification works
We track 11 ETFs representing the main market forces: equities (SPY), value vs. growth (IWD−IWF), momentum (MTUM), quality (QUAL), long-term bonds (TLT), investment grade credit (LQD), high yield (HYG), emerging markets (EEM), volatility (VIXY), commodities (DBC) and gold (GLD). Each week, these assets move together or in opposite directions — and hidden in those patterns are market regimes.
We use Principal Component Analysis (PCA) to extract the four dominant patterns from these 11 assets. Instead of analyzing each ETF separately, PCA finds the "invisible axes" that explain most of the joint movement. These axes are called Principal Components.
Together, these four components explain 78% of all weekly variation across the 11 factors — capturing the essential market dynamics in four complementary dimensions.
On each component, we run a Markov-Switching model that automatically selects the optimal number of regimes (K) by BIC, allowing us to detect regime changes in real time with the ideal granularity for each dimension.
PC1
Risk Appetite
The market thermometer
42% of variance
When this component rises, nearly everything rises together: equities, momentum, quality, credit and EM all move in the same direction, while volatility (VIXY) falls. It's the dominant force — the classic risk-on / risk-off axis.
Risk-On 66%
| ETF | Peso | 1W | 1M |
|---|---|---|---|
| SPY | +0.438 | -0.3% | -6.1% |
| QUAL | +0.431 | -0.4% | -6.3% |
| MTUM | +0.409 | -2.1% | -4.5% |
| HYG | +0.395 | +0.1% | -1.1% |
| EEM | +0.362 | -0.8% | -4.6% |
How to read it
▲ high = risk-on (equities, momentum and credit rallying)
▼ low = risk-off (broad selloff, flight to safety)
▼ low = risk-off (broad selloff, flight to safety)
▶ What each ETF represents
| SPY | S&P 500 total return — broad US equity market exposure |
| QUAL | MSCI USA Quality Factor — stocks with high ROE, stable earnings, low leverage |
| MTUM | MSCI USA Momentum Factor — stocks with strong recent price trends |
| HYG | High-yield corporate bonds (below BBB) — higher credit risk, correlated with equities in stress |
| EEM | iShares MSCI Emerging Markets — broad EM equity exposure (China, Taiwan, India, Korea, Brazil) |
| LQD | Investment-grade corporate bonds (BBB and above) — credit risk with moderate spread |
| DBC | Invesco DB Commodity Index — diversified basket (energy, metals, agriculture) |
| VIXY | ProShares VIX Short-Term Futures — direct proxy for market fear/volatility (VIX) |
| GLD | SPDR Gold Shares — gold price proxy, safe haven and inflation hedge |
| IWD − IWF | Russell 1000 Value minus Growth — spread between value and growth stocks (positive = value outperforms) |
| TLT | 20+ Year US Treasury bonds — long duration, rises when yields fall |
PC2
Duration / Rates
The rates channel
16% of variance
Captures the rates and safe-haven world, independent of risk appetite. TLT, LQD and gold dominate; commodities and equities on the opposite side. Distinguishes panic from rising rates.
Easing 82%
Projection →
Easing
| ETF | Peso | 1W | 1M |
|---|---|---|---|
| TLT | +0.683 | +1.3% | -1.9% |
| LQD | +0.500 | +0.6% | -1.7% |
| GLD | +0.342 | -0.0% | -12.7% |
| VIXY | +0.221 | -0.9% | +10.0% |
| DBC | −0.213 | +0.6% | +6.3% |
How to read it
▲ high = yields falling, bonds rallying (easing or flight-to-quality)
▼ low = yields rising, bonds falling (monetary tightening)
▼ low = yields rising, bonds falling (monetary tightening)
▶ What each ETF represents
| TLT | 20+ Year US Treasury bonds — long duration, rises when yields fall |
| LQD | Investment-grade corporate bonds (BBB and above) — credit risk with moderate spread |
| GLD | SPDR Gold Shares — gold price proxy, safe haven and inflation hedge |
| VIXY | ProShares VIX Short-Term Futures — direct proxy for market fear/volatility (VIX) |
| DBC | Invesco DB Commodity Index — diversified basket (energy, metals, agriculture) |
| SPY | S&P 500 total return — broad US equity market exposure |
| EEM | iShares MSCI Emerging Markets — broad EM equity exposure (China, Taiwan, India, Korea, Brazil) |
| QUAL | MSCI USA Quality Factor — stocks with high ROE, stable earnings, low leverage |
| IWD − IWF | Russell 1000 Value minus Growth — spread between value and growth stocks (positive = value outperforms) |
| HYG | High-yield corporate bonds (below BBB) — higher credit risk, correlated with equities in stress |
| MTUM | MSCI USA Momentum Factor — stocks with strong recent price trends |
PC3
Cyclical Rotation
Value, commodities and cycle
12% of variance
Captures rotation between cyclical assets (value, commodities, gold) and defensive/growth. When it rises, the market favors sectors tied to the economic cycle and inflation.
Cyclical 70%
Projection →
Early-Cycle
| ETF | Peso | 1W | 1M |
|---|---|---|---|
| IWD − IWF | +0.673 | +0.2% | +4.5% |
| DBC | +0.561 | +0.6% | +6.3% |
| GLD | +0.352 | -0.0% | -12.7% |
| QUAL | −0.170 | -0.4% | -6.3% |
| MTUM | −0.148 | -2.1% | -4.5% |
How to read it
▲ high = rotation to value and commodities (cycle expanding, reflation trade)
▼ low = rotation to growth and defensives (cycle slowing)
▼ low = rotation to growth and defensives (cycle slowing)
▶ What each ETF represents
| IWD − IWF | Russell 1000 Value minus Growth — spread between value and growth stocks (positive = value outperforms) |
| DBC | Invesco DB Commodity Index — diversified basket (energy, metals, agriculture) |
| GLD | SPDR Gold Shares — gold price proxy, safe haven and inflation hedge |
| QUAL | MSCI USA Quality Factor — stocks with high ROE, stable earnings, low leverage |
| MTUM | MSCI USA Momentum Factor — stocks with strong recent price trends |
| SPY | S&P 500 total return — broad US equity market exposure |
| HYG | High-yield corporate bonds (below BBB) — higher credit risk, correlated with equities in stress |
| LQD | Investment-grade corporate bonds (BBB and above) — credit risk with moderate spread |
| EEM | iShares MSCI Emerging Markets — broad EM equity exposure (China, Taiwan, India, Korea, Brazil) |
| TLT | 20+ Year US Treasury bonds — long duration, rises when yields fall |
| VIXY | ProShares VIX Short-Term Futures — direct proxy for market fear/volatility (VIX) |
PC4
Tail Risk
Fear and protection
8% of variance
Dominated by volatility (VIXY) and gold — captures fear spikes and demand for protection that don't necessarily show in equity or bond prices.
Alert 95%
Projection →
Alert
| ETF | Peso | 1W | 1M |
|---|---|---|---|
| GLD | +0.545 | -0.0% | -12.7% |
| VIXY | +0.512 | -0.9% | +10.0% |
| IWD − IWF | −0.431 | +0.2% | +4.5% |
| DBC | +0.286 | +0.6% | +6.3% |
| LQD | −0.275 | +0.6% | -1.7% |
How to read it
▲ high = fear spike (VIX rising, gold as refuge, elevated tail risk)
▼ low = calm market (complacency, compressed VIX)
▼ low = calm market (complacency, compressed VIX)
▶ What each ETF represents
| GLD | SPDR Gold Shares — gold price proxy, safe haven and inflation hedge |
| VIXY | ProShares VIX Short-Term Futures — direct proxy for market fear/volatility (VIX) |
| IWD − IWF | Russell 1000 Value minus Growth — spread between value and growth stocks (positive = value outperforms) |
| DBC | Invesco DB Commodity Index — diversified basket (energy, metals, agriculture) |
| LQD | Investment-grade corporate bonds (BBB and above) — credit risk with moderate spread |
| HYG | High-yield corporate bonds (below BBB) — higher credit risk, correlated with equities in stress |
| TLT | 20+ Year US Treasury bonds — long duration, rises when yields fall |
| EEM | iShares MSCI Emerging Markets — broad EM equity exposure (China, Taiwan, India, Korea, Brazil) |
| MTUM | MSCI USA Momentum Factor — stocks with strong recent price trends |
| SPY | S&P 500 total return — broad US equity market exposure |
| QUAL | MSCI USA Quality Factor — stocks with high ROE, stable earnings, low leverage |
Technical details
K=3 Markov-Switching on PC1+PC2 of 11 factor-ETFs (SPY, IWD−IWF, MTUM, QUAL, TLT, LQD, HYG, EEM, VIXY, DBC, GLD)
PCA: PC1 42% + PC2 16% + PC3 12% + PC4 8% = 78% variance · AIC: 2522 · BIC: 2589
PCA: PC1 42% + PC2 16% + PC3 12% + PC4 8% = 78% variance · AIC: 2522 · BIC: 2589
▸Methodological Note — Market Regime Model
What is a market regime?
Financial markets do not behave the same way all the time. There are periods of optimism, where most assets rise in coordinated fashion, and periods of stress, where everything falls together and investors seek protection. Between these extremes, there are transition moments with no clear direction. Identifying which "regime" we are in helps understand the current investment environment.
How does the identification work?
We track 11 asset classes weekly that represent the market's main forces: US equities, value vs. growth, momentum, quality, long-term bonds, corporate credit (high and low quality), emerging markets, volatility, commodities, and gold. Each week, these assets move together or in opposite directions — and in these co-movement patterns lie the regime signals.
What is the Markov-Switching model?
The Markov-Switching model (or regime-switching model) is a statistical technique that assumes the market can be in different "states" and switches between them over time. The name comes from Russian mathematician Andrei Markov, who studied processes where the next state depends only on the current state (not the entire past history).
In practice, the model does the following:
• Assumes distinct states exist (in our case: optimistic, neutral, and stressed)
• In each state, asset returns behave in statistically different ways (different means and volatilities)
• The model calculates, week by week, the probability of being in each state
• When one state's probability exceeds the others, a "regime change" occurs
Why is this useful for investors?
Different asset types perform better in different regimes. Growth stocks tend to shine in optimistic regimes. Gold and government bonds tend to protect in stress regimes. Knowing which regime we are in helps calibrate risk exposure — not as a crystal ball, but as a thermometer of the current situation.
Principal Component Analysis (PCA)
Since there are 11 assets, analyzing them individually would be complex. We use a technique called PCA that extracts the 4 most important movement patterns from these 11 assets. Each pattern (principal component) captures a different market dimension: risk appetite, rates/duration, cyclical rotation, and tail risk. For each dimension, we run a separate Markov-Switching model, allowing a richer and more granular reading of the current regime.
Financial markets do not behave the same way all the time. There are periods of optimism, where most assets rise in coordinated fashion, and periods of stress, where everything falls together and investors seek protection. Between these extremes, there are transition moments with no clear direction. Identifying which "regime" we are in helps understand the current investment environment.
How does the identification work?
We track 11 asset classes weekly that represent the market's main forces: US equities, value vs. growth, momentum, quality, long-term bonds, corporate credit (high and low quality), emerging markets, volatility, commodities, and gold. Each week, these assets move together or in opposite directions — and in these co-movement patterns lie the regime signals.
What is the Markov-Switching model?
The Markov-Switching model (or regime-switching model) is a statistical technique that assumes the market can be in different "states" and switches between them over time. The name comes from Russian mathematician Andrei Markov, who studied processes where the next state depends only on the current state (not the entire past history).
In practice, the model does the following:
• Assumes distinct states exist (in our case: optimistic, neutral, and stressed)
• In each state, asset returns behave in statistically different ways (different means and volatilities)
• The model calculates, week by week, the probability of being in each state
• When one state's probability exceeds the others, a "regime change" occurs
Why is this useful for investors?
Different asset types perform better in different regimes. Growth stocks tend to shine in optimistic regimes. Gold and government bonds tend to protect in stress regimes. Knowing which regime we are in helps calibrate risk exposure — not as a crystal ball, but as a thermometer of the current situation.
Principal Component Analysis (PCA)
Since there are 11 assets, analyzing them individually would be complex. We use a technique called PCA that extracts the 4 most important movement patterns from these 11 assets. Each pattern (principal component) captures a different market dimension: risk appetite, rates/duration, cyclical rotation, and tail risk. For each dimension, we run a separate Markov-Switching model, allowing a richer and more granular reading of the current regime.
Source: EODHD (weekly ETFs), PCA + Markov-Switching
Weekly Reading
The energy sector dominated gains in April with returns of 37.3% for the month and 87.3% for the year, led by Heating Oil (42.6% 1M), WTI Crude (39.9% 1M), and Gas Oil (38.6% 1M), driven by reduced geopolitical tensions in the Middle East following signals of a possible U.S.-Iran agreement that lowered expectations for fuel and transportation costs. In contrast, precious metals faced significant pressure with declines of 13.2% for the month, highlighted by Silver (-16.4% 1M), Palladium (-13.9% 1M), and Gold (-12.4% 1M), while grains posted a moderate gain of 3.2% for the month with wheat hit by technical profit-taking combined with prospects of improved weather in U.S. producing regions. Industrial metals remained virtually stable with a 2.4% drop for the month, reflecting mixed dynamics between improved industrial demand from reduced geopolitical tensions and technical correction pressures. The disconnect between energy's strong rally and falling precious metals suggests portfolio rebalancing where global risk appetite recovers, shifting flows from defensive assets to commodities tied to economic growth.
This panel tracks the performance of major global commodities, their statistical equilibrium relationships, and bilateral trade flows between countries. Together, these indicators reveal supply and demand pressures that affect FX, inflation, and producer stocks.
Commodities — Bloomberg Commodity Indices
Returns panel by category (click to filter). Data from Bloomberg Commodity[?] sub-indices (BCOM). For each commodity, we show the 5 stocks with the highest correlation[?] over the last 30 days.
How to read this panel: Categories are sorted by YTD return (year-to-date). Within each category, each commodity shows returns across different windows (1W, 1M, 3M, YTD). Green = up, red = down. Click a commodity to see the 5 global stocks with the highest correlation over the last 30 days.
▸Methodology Note — Commodities
What is it? The commodities panel shows the recent return of each commodity grouped by category (energy, precious metals, industrial metals, grains, softs, and livestock), using Bloomberg Commodity indices as reference.
How does it work? Returns are calculated from daily closing prices. For each commodity, we identify the 5 global stocks with the highest correlation over the last 30 days — stocks whose prices moved in the same direction and intensity.
Why is it useful? It helps identify which commodities are trending up or down, and which producer or consumer stocks may be affected.
How to read? Categories are sorted by YTD return. Within each category, check returns across different windows (1W, 1M, 3M, YTD). Click a commodity to see the most correlated stocks.
How does it work? Returns are calculated from daily closing prices. For each commodity, we identify the 5 global stocks with the highest correlation over the last 30 days — stocks whose prices moved in the same direction and intensity.
Why is it useful? It helps identify which commodities are trending up or down, and which producer or consumer stocks may be affected.
How to read? Categories are sorted by YTD return. Within each category, check returns across different windows (1W, 1M, 3M, YTD). Click a commodity to see the most correlated stocks.
Source: EODHD — Bloomberg Commodity Indices (BCOM)
Commodity Cointegration — Basket Equilibrium
Monitors historical relationships between commodities using cointegration[?] tests. When two assets that normally move together decouple, the z-score[?] indicates the deviation intensity. The half-life[?] estimates the expected correction time.
▸
Methodology
Methodology: Cointegration tests (Engle-Granger for pairs, Johansen for 3+ assets) identify commodities that maintain a long-run equilibrium relationship. When this relationship breaks (elevated z-score), there is a statistical tendency toward mean reversion. Half-life estimates how long it takes for half the deviation to correct. The simulator lets you project rebalancing scenarios between assets.
Brent vs WTI Engle-Granger · Not cointegrated
Spread between oil benchmarks: reflects logistics and regional dynamics.
Out of Equilibrium
▸
Brent Crude is 4.1% below fair value equilibrium. ~50% of the deviation corrects in 6 days, ~90% corrects in 18 days. Brent Crude tends to rise to rebalance.
Deviation
-3.7σ
Correction in
~6d
Intensity
Extreme
Cointegration vector
1.00×Brent Crude(762.5)−5.86×WTI Crude Oil(141.0)
The σ deviation measures the statistical displacement of this vector from its historical mean — it is not the market spread.
Equilibrium Simulator
If Brent Crude changes by
%
Enter a % change to simulate
Crack Spread Johansen · Cointegrated
Refining margin: relationship between crude oil and its products (gasoline and heating oil).
Out of Equilibrium
▸
Heating Oil outperforming basket (+112.9% 60d). ~50% of the deviation corrects in 29 days, ~90% corrects in 95 days. Heating Oil tends to fall to rebalance.
Deviation
+2.4σ
Correction in
~29d
Intensity
High
Cointegration vector
1.00×WTI Crude Oil(141.0)−0.21×Gasoline (RBOB)(742.8)+0.05×Heating Oil(1133.9)
The σ deviation measures the statistical displacement of this vector from its historical mean — it is not the market spread.
Equilibrium Simulator
If
changes by
%
Enter a % change to simulate
Crude Oil vs Natural Gas Engle-Granger · Not cointegrated
Energy substitution: crude oil and natural gas as alternative sources.
Out of Equilibrium
▸
WTI Crude Oil is 43.4% above fair value equilibrium. ~50% of the deviation corrects in 27 days, ~90% corrects in 88 days. WTI Crude Oil tends to fall to rebalance.
Deviation
+2.2σ
Correction in
~27d
Intensity
High
Cointegration vector
1.00×WTI Crude Oil(141.0)+0.09×Natural Gas(28.9)
The σ deviation measures the statistical displacement of this vector from its historical mean — it is not the market spread.
Equilibrium Simulator
If WTI Crude Oil changes by
%
Enter a % change to simulate
Base Metals Johansen · Not cointegrated
Industrial metals sensitive to the global economic cycle and Chinese demand.
Warning
▸
Aluminum outperforming basket (+10.2% 60d). ~50% of the deviation corrects in 10 days, ~90% corrects in 32 days. Aluminum tends to fall to rebalance.
Deviation
-1.6σ
Correction in
~10d
Intensity
Moderate
Cointegration vector
1.00×Copper(434.5)−19.92×Aluminum(39.7)+5.94×Zinc(98.0)
The σ deviation measures the statistical displacement of this vector from its historical mean — it is not the market spread.
Equilibrium Simulator
If
changes by
%
Enter a % change to simulate
Feed → Protein Chain Johansen · Not cointegrated
Feed cost (corn) vs. protein price (cattle and hogs). Rancher margin.
In Equilibrium
▸
Feed cost (corn) vs. protein price (cattle and hogs). Rancher margin.
Deviation
+1.3σ
Correction in
~22d
Intensity
Normal
Cointegration vector
1.00×Corn(8.0)+0.09×Live Cattle(230.4)−13.02×Lean Hogs(4.0)
The σ deviation measures the statistical displacement of this vector from its historical mean — it is not the market spread.
Equilibrium Simulator
If
changes by
%
Enter a % change to simulate
Soybean Crush Johansen · Not cointegrated
Crush margin: raw soybeans vs. by-products (oil and meal).
In Equilibrium
▸
Crush margin: raw soybeans vs. by-products (oil and meal).
Deviation
+1.3σ
Correction in
~37d
Intensity
Normal
Cointegration vector
1.00×Soybeans(255.8)+7.87×Soybean Oil(83.2)+0.68×Soybean Meal(626.5)
The σ deviation measures the statistical displacement of this vector from its historical mean — it is not the market spread.
Equilibrium Simulator
If
changes by
%
Enter a % change to simulate
Kansas vs Chicago Wheat Engle-Granger · Cointegrated
Quality spread: hard red winter (Kansas) vs. soft red winter (Chicago) wheat.
In Equilibrium
▸
Kansas Wheat (HRW) is 1.6% above fair value equilibrium.
Deviation
+1.1σ
Correction in
~12d
Intensity
Normal
Cointegration vector
1.00×Kansas Wheat (HRW)(20.0)−4.80×Chicago Wheat (SRW)(4.0)
The σ deviation measures the statistical displacement of this vector from its historical mean — it is not the market spread.
Equilibrium Simulator
If Kansas Wheat (HRW) changes by
%
Enter a % change to simulate
Livestock Johansen · Not cointegrated
Animal protein chain: feeder cattle, live cattle, and lean hogs.
In Equilibrium
▸
Animal protein chain: feeder cattle, live cattle, and lean hogs.
Deviation
-1.1σ
Correction in
~24d
Intensity
Normal
Cointegration vector
1.00×Live Cattle(230.4)−3.70×Feeder Cattle(191.7)+592.46×Lean Hogs(4.0)
The σ deviation measures the statistical displacement of this vector from its historical mean — it is not the market spread.
Equilibrium Simulator
If
changes by
%
Enter a % change to simulate
Copper vs Gold Engle-Granger · Cointegrated
Copper/gold ratio: classic risk-appetite gauge. Divergences signal regime change.
In Equilibrium
▸
Copper is 3.1% below fair value equilibrium.
Deviation
-1.0σ
Correction in
~12d
Intensity
Normal
Cointegration vector
1.00×Copper(434.5)−0.41×Gold(403.9)
The σ deviation measures the statistical displacement of this vector from its historical mean — it is not the market spread.
Equilibrium Simulator
If Copper changes by
%
Enter a % change to simulate
Tropical Softs Johansen · Not cointegrated
Tropical agricultural commodities sensitive to weather and producer-country FX.
In Equilibrium
▸
Tropical agricultural commodities sensitive to weather and producer-country FX.
Deviation
+0.9σ
Correction in
~45d
Intensity
Normal
Cointegration vector
1.00×Sugar(96.2)+1.42×Coffee(24.0)+0.03×Cocoa(123.1)
The σ deviation measures the statistical displacement of this vector from its historical mean — it is not the market spread.
Equilibrium Simulator
If
changes by
%
Enter a % change to simulate
Ethanol Chain Johansen · Not cointegrated
Ethanol feedstocks (sugar + corn) vs. crude oil — agricultural energy arbitrage.
In Equilibrium
▸
Ethanol feedstocks (sugar + corn) vs. crude oil — agricultural energy arbitrage.
Deviation
-0.6σ
Correction in
~33d
Intensity
Normal
Cointegration vector
1.00×Sugar(96.2)−14.86×Corn(8.0)−0.15×WTI Crude Oil(141.0)
The σ deviation measures the statistical displacement of this vector from its historical mean — it is not the market spread.
Equilibrium Simulator
If
changes by
%
Enter a % change to simulate
Soybean Oil vs Crude Oil Engle-Granger · Not cointegrated
Biodiesel link: soybean oil as biofuel feedstock competes with crude oil.
In Equilibrium
▸
Soybean Oil is 2.2% below fair value equilibrium.
Deviation
-0.6σ
Correction in
~42d
Intensity
Normal
Cointegration vector
1.00×Soybean Oil(83.2)−0.30×WTI Crude Oil(141.0)
The σ deviation measures the statistical displacement of this vector from its historical mean — it is not the market spread.
Equilibrium Simulator
If Soybean Oil changes by
%
Enter a % change to simulate
Gold & Silver Engle-Granger · Not cointegrated
Gold/silver ratio: both are stores of value, but silver has an industrial component.
In Equilibrium
▸
Gold is 4.2% above fair value equilibrium.
Deviation
+0.5σ
Correction in
~30d
Intensity
Normal
Cointegration vector
1.00×Gold(403.9)−0.49×Silver(518.6)
The σ deviation measures the statistical displacement of this vector from its historical mean — it is not the market spread.
Equilibrium Simulator
If Gold changes by
%
Enter a % change to simulate
Platinum vs Gold Engle-Granger · Not cointegrated
Platinum has industrial use (catalysts); gold is a store of value. Divergence signals industrial cycle.
In Equilibrium
▸
Platinum is 0.7% above fair value equilibrium.
Deviation
+0.1σ
Correction in
~28d
Intensity
Normal
Cointegration vector
1.00×Platinum(982.2)−4.75×Gold(403.9)
The σ deviation measures the statistical displacement of this vector from its historical mean — it is not the market spread.
Equilibrium Simulator
If Platinum changes by
%
Enter a % change to simulate
Broad Soft Commodities Johansen · Cointegrated
Broad soft commodities — sensitive to weather and producer-country FX.
In Equilibrium
▸
Broad soft commodities — sensitive to weather and producer-country FX.
Deviation
-0.0σ
Correction in
~24d
Intensity
Normal
Cointegration vector
1.00×Sugar(96.2)−3.80×Coffee(24.0)−0.32×Orange Juice(35.1)−11.39×Cotton(21.1)
The σ deviation measures the statistical displacement of this vector from its historical mean — it is not the market spread.
Equilibrium Simulator
If
changes by
%
Enter a % change to simulate
Grains Johansen · Not cointegrated
Grains compete for acreage and serve as animal feed — tend to move together.
In Equilibrium
▸
Grains compete for acreage and serve as animal feed — tend to move together.
Deviation
+0.0σ
Correction in
~28d
Intensity
Normal
Cointegration vector
1.00×Corn(8.0)−1.73×Wheat(4.0)+0.05×Soybeans(255.8)
The σ deviation measures the statistical displacement of this vector from its historical mean — it is not the market spread.
Equilibrium Simulator
If
changes by
%
Enter a % change to simulate
How to read these cards: Each card is a basket of cointegrated commodities. The status badge shows whether the relationship is in equilibrium (green), warning (yellow), or out of equilibrium (red). The z-score measures how many standard deviations the current price is from the historical mean — above ±2σ is considered extreme. The half-life estimates how many days correction takes. Use the simulator to project rebalancing scenarios.
Source: EODHD commodities.db — Engle-Granger / Johansen
Global Trade Flow Map
Visualization of major bilateral trade[?] corridors, 2014–2025. Gold nodes are net exporters; blue are net importers. Data: UN Comtrade[?].
MTM Exposure — Country Risk
Estimated impact of recent commodity returns on trade balances. Click a country for details. MTM = net balance × price change.
Countries colored by commodity MTM (mark-to-market) impact on trade balance. Red = loss (net importer when price rises, or exporter when it falls). Green = gain. Toggle between % of GDP and absolute value. Click a country for the commodity breakdown.
Chokepoint Simulation
Simulate the impact of closures at major maritime straits and canals on global trade across ALL commodities. See total value at risk and the most vulnerable countries (Top 5 affected exporters and importers).
Closure Control
▸Methodology Note — Trade Flow
What is it? An interactive map of the largest bilateral commodity trade corridors, based on official UN data (UN Comtrade).
How does it work? For each selected commodity, the map shows the largest export and import flows between countries. Curved lines represent trade routes — thicker lines indicate higher traded value. Gold nodes are net exporters; blue nodes are net importers.
Why is it useful? It reveals trade dependencies between countries and how shocks to a producer (crop, sanctions, logistics) can affect global prices.
How to read? Select the commodity, exporter, and importer from the menus. Use the year buttons to compare evolution. Click a country to see origin and destination details.
How does it work? For each selected commodity, the map shows the largest export and import flows between countries. Curved lines represent trade routes — thicker lines indicate higher traded value. Gold nodes are net exporters; blue nodes are net importers.
Why is it useful? It reveals trade dependencies between countries and how shocks to a producer (crop, sanctions, logistics) can affect global prices.
How to read? Select the commodity, exporter, and importer from the menus. Use the year buttons to compare evolution. Click a country to see origin and destination details.
Source: UN Comtrade (bilateral trade, 2014–2025)
Weekly Reading
With dashboard data unavailable, the DI curve reflects a high-interest-rate environment, aligned with Focus expectations projecting Selic at 15% at the end of the Copom tightening cycle, following February inflation data with IPCA at 1.2%, while implied inflation in the Anbima ETTJ hovers around IPCA+5.5% to 6% p.a. for long-term bonds like NTN-B 2026-2045. This setup compresses spreads in fixed income, but selected IPCA+ bonds show attractive spreads of up to 2% versus ETTJ in recent issuances, driven by the Copom's decision to maintain monetary tightening amid persistent fiscal risks. The recent fiscal filing for pension funds reinforces actuarial targets at IPCA+5.6% p.a., signaling potential in assets yielding above the curve.
This panel covers the Brazilian fixed income market — government bonds, yield curves, market expectations, and stochastic simulations. It helps evaluate bond opportunities, track inflation and rate expectations, and understand the term structure.
Fixed Income
How much do government bonds yield today — and are they paying above or below fair value? The table compares each IPCA+[?] bond's real rate with the theoretical ETTJ[?] curve from ANBIMA. Positive spreads indicate opportunity — the bond pays above the curve. Compare Monte Carlo scenarios with CDI[?] returns.
Return Scenarios — Active Bonds
Ref: 2026-03-30| Bond | Current Rate [?] | MC Focus [?] | MC Adjusted [?] |
|---|---|---|---|
| IPCA+ 2026 | 8.72% | 13.51% [13.1 — 13.9] |
13.65% [13.2 — 14.1] |
| IPCA+ 2029 | 8.04% | 12.42% [12.0 — 12.8] |
13.19% [12.8 — 13.6] |
| IPCA+ 2032 | 7.91% | 12.15% [11.8 — 12.5] |
13.08% [12.8 — 13.4] |
| IPCA+ 2035 | 7.63% | 11.83% [11.6 — 12.1] |
12.79% [12.5 — 13.1] |
| IPCA+ 2040 | 7.29% | 11.44% [11.2 — 11.7] |
12.44% [12.2 — 12.7] |
| IPCA+ 2045 | 7.18% | 11.32% [11.1 — 11.5] |
12.33% [12.1 — 12.5] |
| IPCA+ 2050 | 7.13% | 11.26% [11.1 — 11.4] |
12.29% [12.1 — 12.4] |
| Prefixado 2027 | 14.09% | 14.09% | 14.09% |
| Prefixado 2028 | 14.24% | 14.24% | 14.24% |
| Prefixado 2029 | 14.25% | 14.25% | 14.25% |
| Prefixado 2031 | 14.32% | 14.32% | 14.32% |
| Prefixado 2032 | 14.38% | 14.38% | 14.38% |
| Selic 2027 | 0.00% | 13.31% [12.4 — 14.2] |
13.35% [12.5 — 14.2] |
| Selic 2028 | 0.01% | 12.15% [11.2 — 13.2] |
12.48% [11.5 — 13.5] |
| Selic 2029 | 0.04% | 11.61% [10.7 — 12.4] |
12.12% [11.2 — 13.0] |
| Selic 2031 | 0.09% | 11.20% [10.6 — 11.8] |
11.85% [11.2 — 12.4] |
How to read: Estimated annualised nominal IRR via Monte Carlo (5,000 simulations). Rates for each scenario (IPCA or Selic) are compounded month by month along the simulated trajectory up to the bond's maturity — not a flat rate, but the actual accumulation along the projected path. Long bonds (IPCA+ 2040, 2045, 2050) are simulated over their full maturity, with mean-reversion around long-run Focus targets.
MC Focus ■: uses Focus projections as published by BCB (market consensus). MC Adjusted ■: adjusts targets for Focus systematic historical bias (IPCA: -0.55pp@12M / -0.99pp@24M, Selic: -0.06pp@12M / -0.85pp@24M — 2016–2025). Bias horizon-matched: 12M for current year, 24M for next year. Negative bias means Focus tends to underestimate realized values.
IPCA+: IRR = (1 + locked real rate) × (1 + simulated cumulative IPCA) − 1. Selic: IRR = simulated compounded CDI + spread. Prefixado: rate locked at purchase (no uncertainty). [P25–P75] = interquartile range — 50% of simulations fall within this interval.
MC Focus ■: uses Focus projections as published by BCB (market consensus). MC Adjusted ■: adjusts targets for Focus systematic historical bias (IPCA: -0.55pp@12M / -0.99pp@24M, Selic: -0.06pp@12M / -0.85pp@24M — 2016–2025). Bias horizon-matched: 12M for current year, 24M for next year. Negative bias means Focus tends to underestimate realized values.
IPCA+: IRR = (1 + locked real rate) × (1 + simulated cumulative IPCA) − 1. Selic: IRR = simulated compounded CDI + spread. Prefixado: rate locked at purchase (no uncertainty). [P25–P75] = interquartile range — 50% of simulations fall within this interval.
Historical Simulator
Updated: 2026-03-27
Entry Date:
Early Exit:
Off
▸Methodology Note — Fixed Income
What is it? An integrated view of the Brazilian government bond market. It combines actual Tesouro Direto prices, the ANBIMA-estimated term structure (ETTJ), B3-traded futures curves, and the Central Bank's Focus Survey market projections.
How does it work?
• ETTJ Table: Compares each IPCA+ bond's real rate with the theoretical ANBIMA curve, calculating the spread in basis points and projected IRR.
• B3 Curves: Shows DI futures (nominal interest rate) and FX-hedged rate (FRC) curves, extracted daily from B3.
• ANBIMA ETTJ: Term structure estimated via Svensson model for 13 maturities (1M to 15Y), decomposed into nominal rate, real rate, and break-even inflation.
• Focus: Market median projections for 11 macro indicators, with historical accuracy analysis.
• Monte Carlo: Stochastic simulations of future IPCA and Selic paths using Vasicek and Brownian Bridge models, calibrated with Focus data and DI curve.
Why is it useful? It helps identify bonds trading above fair value (positive spread vs ETTJ), understand market expectations for rates and inflation, and simulate probabilistic scenarios.
How to read? In the table, positive spreads (green) indicate the bond offers a rate above the theoretical curve. In the curves, compare slopes to assess expectations for rate increases or decreases. In Focus, watch the direction of revision arrows.
How does it work?
• ETTJ Table: Compares each IPCA+ bond's real rate with the theoretical ANBIMA curve, calculating the spread in basis points and projected IRR.
• B3 Curves: Shows DI futures (nominal interest rate) and FX-hedged rate (FRC) curves, extracted daily from B3.
• ANBIMA ETTJ: Term structure estimated via Svensson model for 13 maturities (1M to 15Y), decomposed into nominal rate, real rate, and break-even inflation.
• Focus: Market median projections for 11 macro indicators, with historical accuracy analysis.
• Monte Carlo: Stochastic simulations of future IPCA and Selic paths using Vasicek and Brownian Bridge models, calibrated with Focus data and DI curve.
Why is it useful? It helps identify bonds trading above fair value (positive spread vs ETTJ), understand market expectations for rates and inflation, and simulate probabilistic scenarios.
How to read? In the table, positive spreads (green) indicate the bond offers a rate above the theoretical curve. In the curves, compare slopes to assess expectations for rate increases or decreases. In Focus, watch the direction of revision arrows.
Source: Tesouro Direto, ANBIMA (ETTJ), BCB SGS (IPCA, CDI)
Updated: 2026-03-31
Weekly change (bps) — Prefixado vs 2026-03-23
| 1A | 2A | 5A | 10A | 15A |
|---|---|---|---|---|
| +24 | +31 | +30 | +27 | +24 |
How to read these curves: The X axis is maturity; Y is annualized rate. The DI curve shows the nominal interest rate the market expects for each maturity. An upward-sloping curve indicates expectations of rising rates; downward, falling. The FX-hedged rate (FRC) reflects the cost of FX hedging — the DI-FRC gap indicates currency risk premium.
▸Methodology Note — Yield Curves
What is it? Yield curves show the rate the market expects for each maturity. Two sets:
• B3 Curves: Extracted from futures contracts traded on B3 — DI1 reflects the expected nominal interest rate, and FRC (FX-hedged rate) reflects the cost of FX hedging in USD.
• ANBIMA ETTJ: Theoretical curves estimated by ANBIMA using the Svensson model (6 parameters) for 13 maturities (1 month to 15 years). Decomposed into: nominal rate (Prefixado), real rate (IPCA+), and break-even inflation (difference between the two).
Why is it useful? Curve slope reveals expectations: an upward-sloping curve suggests the market expects higher future rates; inverted, a decrease. Dashed curves show the previous week for comparison — shifts indicate recent changes in expectations.
How to read? Compare solid curves (current) with dashed (previous week). If the solid curve is above the dashed, rates have opened (market more pessimistic on rates). Break-even inflation (yellow) is the difference between Prefixado and IPCA+ — shows how much inflation the market prices for each maturity.
• B3 Curves: Extracted from futures contracts traded on B3 — DI1 reflects the expected nominal interest rate, and FRC (FX-hedged rate) reflects the cost of FX hedging in USD.
• ANBIMA ETTJ: Theoretical curves estimated by ANBIMA using the Svensson model (6 parameters) for 13 maturities (1 month to 15 years). Decomposed into: nominal rate (Prefixado), real rate (IPCA+), and break-even inflation (difference between the two).
Why is it useful? Curve slope reveals expectations: an upward-sloping curve suggests the market expects higher future rates; inverted, a decrease. Dashed curves show the previous week for comparison — shifts indicate recent changes in expectations.
How to read? Compare solid curves (current) with dashed (previous week). If the solid curve is above the dashed, rates have opened (market more pessimistic on rates). Break-even inflation (yellow) is the difference between Prefixado and IPCA+ — shows how much inflation the market prices for each maturity.
Source: B3 Derivatives (DI1, FRC)
Source: ANBIMA via pyettj (Svensson model)
Focus Survey — Market Expectations
| Indicator | 2026 | 2027 | ||
|---|---|---|---|---|
| Median | Trend | Median | Trend | |
| IPCA | 4.31% [3.40 — 5.68] |
3.84% [3.00 — 6.00] |
||
| Selic | 12.50% a.a. [11.00 — 14.75] |
10.50% a.a. [8.00 — 14.75] |
||
| FX Rate (BRL/USD) | 5.40 [4.90 — 6.00] |
5.45 [4.50 — 6.00] |
||
| GDP | 1.85% [1.19 — 2.47] |
1.80% [1.00 — 2.80] |
||
| IGP-M | 3.46% [1.90 — 5.78] |
4.00% [1.96 — 6.52] |
||
| Gross Debt / GDP | 83.60% PIB [79.00 — 86.55] |
86.85% PIB [80.00 — 92.14] |
||
| Primary Balance / GDP | -0.50% PIB [-1.00 — 0.20] |
-0.40% PIB [-1.13 — 0.50] |
||
| IPCA Administered | 4.27% [2.65 — 6.40] |
3.77% [2.47 — 5.89] |
||
| IPCA Services | 5.42% [3.60 — 6.91] |
4.78% [2.68 — 6.36] |
||
| IPCA Market Prices | 4.29% [2.80 — 5.81] |
3.93% [2.17 — 5.22] |
||
| Unemployment | 5.60% [4.80 — 6.61] |
6.10% [4.80 — 7.89] |
||
Source: BCB / Focus Survey (2026-03-31)
Focus Survey — Historical Error & Bias (2016–2025)
| Indicator | 6M MAE | 12M MAE | 24M MAE |
|---|---|---|---|
| IPCA |
1.36
bias -0.3 · n=10
|
1.38 ▼
bias -0.6 · n=10
|
1.59 ▼
bias -1.0 · n=10
|
| Selic |
0.85
bias -0.1 · n=10
|
2.29
bias -0.1 · n=10
|
4.55 ▼
bias -0.8 · n=10
|
| FX Rate |
0.27
bias -0.1 · n=10
|
0.61
bias -0.1 · n=10
|
0.76 ▼
bias -0.5 · n=10
|
| GDP |
0.98 ▼
bias -0.9 · n=10
|
1.86
bias -0.2 · n=10
|
2.07 ▲
bias +0.5 · n=10
|
| IGP-M |
3.84 ▼
bias -1.5 · n=10
|
5.81 ▼
bias -3.1 · n=10
|
6.14 ▼
bias -3.6 · n=10
|
| Unemployment |
1.44 ▲
bias +1.4 · n=4
|
2.20 ▲
bias +2.2 · n=4
|
3.36 ▲
bias +3.4 · n=3
|
MAE = mean absolute error. Bias: ▲ = overestimates, ▼ = underestimates (|bias| > 0.3)
How to read this table: Each row is a macro indicator (Selic, IPCA, GDP, etc.) with the market median projection for this year and next. Trend arrows (▲ / ▼) show whether projections are being revised up or down in recent weeks. Sparklines show the evolution of projections over time.
▸Methodology Note — Focus Survey
What is it? The Focus Survey is a weekly poll by Brazil's Central Bank collecting projections from ~130 financial institutions for key macroeconomic indicators: Selic, IPCA, GDP, FX, trade balance, and others.
How does it work? Every Friday the BCB publishes the projection medians for the current and next year. The table shows these medians along with sparklines revealing the recent revision trend. Arrows indicate whether projections are being revised up or down.
Accuracy Analysis: Below the table, we analyze Focus's track record since 2016 — measuring mean absolute error (MAE), bias (whether the market tends to be optimistic or pessimistic), and how accuracy varies with horizon (December projections are more precise than January ones).
Why is it useful? Shows market consensus — and whether that consensus is being revised. When many projections shift in the same direction, it may signal a changing macro outlook.
How does it work? Every Friday the BCB publishes the projection medians for the current and next year. The table shows these medians along with sparklines revealing the recent revision trend. Arrows indicate whether projections are being revised up or down.
Accuracy Analysis: Below the table, we analyze Focus's track record since 2016 — measuring mean absolute error (MAE), bias (whether the market tends to be optimistic or pessimistic), and how accuracy varies with horizon (December projections are more precise than January ones).
Why is it useful? Shows market consensus — and whether that consensus is being revised. When many projections shift in the same direction, it may signal a changing macro outlook.
▸Methodology Note — Monte Carlo Simulations
What is it? Monte Carlo simulation generates thousands of possible paths for an indicator, allowing you to visualize the distribution of future scenarios instead of a single point forecast.
How does it work? Two distinct models:
• IPCA (Vasicek): Mean-reverting process — inflation tends to converge to the Focus target, with speed calibrated by historical persistence. Volatility is estimated from past Focus forecast errors.
• Selic (Brownian Bridge): Path guided by the B3 DI1 futures curve as a "backbone", connecting the current value to the Focus target. Uncertainty grows then shrinks approaching the anchor point.
Why is it useful? Instead of asking "what will Selic be?", it shows "what is the probability of Selic being above X%?". Allows assessing tail risks and extreme scenarios.
How to read? The dark band (P25–P75) covers the 50% most likely scenarios. The light band (P5–P95) covers 90% of scenarios. The center line is the median. The probability card summarizes the chance of exceeding a specific threshold.
How does it work? Two distinct models:
• IPCA (Vasicek): Mean-reverting process — inflation tends to converge to the Focus target, with speed calibrated by historical persistence. Volatility is estimated from past Focus forecast errors.
• Selic (Brownian Bridge): Path guided by the B3 DI1 futures curve as a "backbone", connecting the current value to the Focus target. Uncertainty grows then shrinks approaching the anchor point.
Why is it useful? Instead of asking "what will Selic be?", it shows "what is the probability of Selic being above X%?". Allows assessing tail risks and extreme scenarios.
How to read? The dark band (P25–P75) covers the 50% most likely scenarios. The light band (P5–P95) covers 90% of scenarios. The center line is the median. The probability card summarizes the chance of exceeding a specific threshold.
P(IPCA > 4.5%) in 2027
61%
Final Distribution (2027)
P5 2.8% ·
P50 4.9% ·
P95 6.8%
Parameters: κ=2.0 · σ=2.35% p.a. · bias: -0.55pp@12M / -0.99pp@24M · 5,000 simulations · 24 months
σ from Focus 12M historical error · horizon-matched bias (12M current year, 24M next year)
σ from Focus 12M historical error · horizon-matched bias (12M current year, 24M next year)
How to read this chart: The "fan" chart shows 1,000 simulated paths. The darker band is the P25–P75 range (most likely scenario); the lighter band is P5–P95 (extreme scenarios). The center line is the median (P50). The card on the left shows the probability of the indicator exceeding a given threshold.
P(Selic > 10%) in 2027
67%
Final Distribution (2027)
P5 6.3% ·
P50 11.3% ·
P95 16.3%
Parameters: σ=3.00% p.a. · bias: -0.06pp@12M / -0.85pp@24M · 5,000 simulations · 20 months
σ from Focus 12M historical error · horizon-matched bias (12M current year, 24M next year) · Brownian bridge over DI1
σ from Focus 12M historical error · horizon-matched bias (12M current year, 24M next year) · Brownian bridge over DI1
How to read this chart: The "fan" chart shows 1,000 simulated paths. The darker band is the P25–P75 range (most likely scenario); the lighter band is P5–P95 (extreme scenarios). The center line is the median (P50). The card on the left shows the probability of the indicator exceeding a given threshold.
Source: BCB Focus (targets), B3 DI1 (curve), historical Focus errors (volatility)
Weekly Analysis 31/03/2026 14:53
The geopolitical sentiment of the week, measured by the VADER score on headlines, reveals extreme tensions in the Middle East, with Lebanon (-0.823), Israel (-0.681), Yemen (-0.502) and Iran (-0.490) recording the worst negative indicators, driven by escalations such as Israeli attacks on Iran and Lebanon on March 24, a missile injuring four in Tel Aviv and Iran setting fire to an oil tanker in Dubai on March 30, in addition to threats of closing the Strait of Hormuz. Countries like Ukraine (+0.402) and France (+0.402) show positive scores, with Macron on a trip to Asia to strengthen Indo-Pacific ties independent of China and the US. These events connect directly to the surge in oil volatility (OVX current 96.6, +199.9% vs 3M ago) and drops in stock markets like Peru (1M -18.9%) and South Africa (1M -21.3%), reflecting global energy risks.
In the global indices map, Norway leads with 1M return of +9.9% and YTD +25.2%, supported by GDP of 2.15%, controlled CPI at 2.47% and nearly flat curve (-0.01), while Nigeria surprises with YTD +28.8% despite explosive CPI at 33.24%. Emerging regions like EMEA and LatAm show divergences: Russia (YTD +1.2%, GDP 3.83%) resists, but Argentina stagnates at 0.0% with negative GDP (-1.68%) and CPI of 219.88%; Romania (YTD +13.6%) grows despite GDP -1.46%. At the bottom, Asia and LatAm suffer: Vietnam (-11.7% 1M), Indonesia (-15.7%) and Peru (-18.9%), even with solid GDPs like 6.11% in Vietnam, highlighting external pressures on fundamentals.
Global FX points to moderate appreciations in emergings like Nigerian naira (3M +4.1%), Brazilian real (3M +4.1%) and Argentine peso (3M +3.7%), contrasting with depreciations in Asia and EMEA: Korean won (1M -6.1%), Thai baht (-5.8%) and Egyptian pound (-11.6% 1M). The FX-stock loadings reveal vulnerabilities: South Africa (loading -1.677, p=0.000), Peru (-1.374) and Chile (-1.153) are the most exposed, where depreciations drag down local indices, as seen in the 1M of -21.3% in ZAF and -18.9% in Peru; resilient ones like New Zealand (+0.194, p=0.002) and Australia (+0.248) mitigate impacts, aligning with steepened curves like NZ (+1.96%).
Risk perception rises with volatility exploding: VIX at 31.1 (+116.7% vs 3M), VXEEM at 39.2 (+143.8%) for emergings, GVZ gold 45.5 (+76.5%) and OVX oil 96.6 (+199.9%), signaling widespread aversion. Credit spreads widen in all categories: IG AAA from 0.35% to 0.42% (+0.07%), HY CCC from 8.88% to 10.13% (+1.25%), with high yield B (+0.71%) and BB (+0.47%) more sensitive, indicating risk appetite in retraction and greater default pricing in lower quality paper.
Geopolitically, the negative scores in ISR, IRN and LBN connect to drops in commodity and exposed stock volatility: OVX +199.9% reflects Iranian attacks on tankers and Trump threats to Iranian facilities on March 30, impacting Peru (sentiment +0.382 but index 1M -18.9% and loading -1.374) and Japan (1M -12.8%, neutral sentiment -0.028 with oil pressures). Russia (sentiment -0.440) and India (-0.440), affected by expiration of Russian oil authorizations, see weak currencies like Indian rupee (3M -4.9%), while China (+0.389) resists with CSI 300 +0.6%.
Connections between data reinforce the picture: VIX +116.7% and VXEEM +143.8% coincide with HY spreads widening +1.25% in CCC and depreciations in EM like Peru and Indonesia (loading -0.843), amplifying drops in bottom indices like ZAF -21.3% 1M; Brazil stands out with real +4.1% 3M, real interest +9.2% (interest 14.75%, CPI 5.53%) and inverted curve (-1.59%), but unlisted loading suggests relative resilience. Small caps vol RVX +86.3% and credit IG BBB +0.13% point to systemic risk, while leaders like Norway deviate with positive GDP and low inflation.