Regime MonitorReading the market regime through cross-asset correlation, currencies and rates — a read-only snapshot built on daily closes
Each card shows the current value, the latest observation date, where it sits in its distribution, and the series itself. Missing values are left empty.
Cross-asset synchronization
Observed 2026-07-27- 60-day window
- 0.4084.7 percentile
- All pairs (reference)
- 0.3430.0 percentile
- 1Y median
- 0.26
The average correlation across pairs drawn from different asset classes. The higher it runs, the less diversification is working.
Downside correlation
Observed 2026-07-27No time series
- Up-day correlation
- 0.21
- Down minus up
- -0.05
- Down-day sample
- 56d
- Up-day sample
- 64d
Correlation computed only on days the S&P fell, placed next to the up-day figure. The gap between them is how much truth there is in 'everything correlates on the way down'.
Korean won vs dollar index
Observed 2026-07-27- 60-day correlation
- 0.02
- Yen vs dollar index
- 0.1176.8 percentile
- Won vs yen
- 0.5368.3 percentile
When the dollar strengthens and the won weakens with it, the correlation is positive. There is no assumed 'normal' level — only where it sits in its own distribution.
Nasdaq rate beta
Observed 2026-07-27- r-squared
- 0.219
- Standard error
- 0.0356
- Momentum factor (MTUM)
- -1.83%16.5 percentile
The Nasdaq's same-day return per 10bp move in the 10-year yield. The more negative it is, the more rates are steering equities.
Treasury yield levels
Observed 2026-07-27No time series
- 1-day move
- -3.8bp
- 5-day move
- +4.3bp
- 20-day move
- +26.9bp
- 30-year Treasury
- 5.13%97.0 percentile
Closing yields on the 10-year and 30-year Treasury plus recent moves. Percentiles are positions within the last year's distribution.
Risk-asset 20-day momentum
Observed 2026-07-27No time series
- Semiconductors
- -10.3%
- Software
- +3.1%
- Nasdaq
- -1.4%
- Korea
- -18.3%
The average 20-session return across semiconductors, software, the Nasdaq and Korea. This is what sets the direction of the verdict.
Correlation matrix
Higher values mean the two moved in the same direction. Blue means they moved opposite.| Software | Semiconductors | S&P 500 | Nasdaq | Korea | Gold | Copper | Bitcoin | Long Treasuries | |
|---|---|---|---|---|---|---|---|---|---|
| Software | — | -0.14 | 0.41 | 0.40 | -0.13 | 0.34 | 0.23 | 0.36 | 0.02 |
| Semiconductors | -0.14 | — | 0.69 | 0.76 | 0.88 | 0.25 | 0.40 | 0.20 | -0.03 |
| S&P 500 | 0.41 | 0.69 | — | 0.96 | 0.61 | 0.53 | 0.49 | 0.44 | 0.03 |
| Nasdaq | 0.40 | 0.76 | 0.96 | — | 0.67 | 0.49 | 0.43 | 0.42 | 0.05 |
| Korea | -0.13 | 0.88 | 0.61 | 0.67 | — | 0.37 | 0.35 | 0.33 | 0.16 |
| Gold | 0.34 | 0.25 | 0.53 | 0.49 | 0.37 | — | 0.57 | 0.68 | -0.05 |
| Copper | 0.23 | 0.40 | 0.49 | 0.43 | 0.35 | 0.57 | — | 0.42 | -0.30 |
| Bitcoin | 0.36 | 0.20 | 0.44 | 0.42 | 0.33 | 0.68 | 0.42 | — | -0.12 |
| Long Treasuries | 0.02 | -0.03 | 0.03 | 0.05 | 0.16 | -0.05 | -0.30 | -0.12 | — |
Key pairs — 30-day correlation
The number on the right is where the pair sits in its own one-year distribution.
- Correlation is not causation. High synchronization only says assets moved together, not which one pulled the other.
- Equity indices are structurally correlated with one another, so the headline synchronization index averages only pairs drawn from different asset classes.
- Percentiles are measured against the most recent 252 trading days. Earlier regimes are not in the comparison set.
- Downside correlation uses only the days the S&P fell within the last 120 trading days.
- The rate beta is a 60-day sample regression. Read it together with the standard error and r-squared.
- The trading calendar follows the S&P, so days when only FX or crypto traded are excluded.
- Everything is daily closing data. Intraday swings show up in the next day's value.
A read-only snapshot computed from daily closing prices, not live data. No weights, targets, or probabilities.
Price data source: yfinance 일봉 종가 (auto_adjust=True, period=2y)