Correlation
of values
-37%
In sync
of periods
48%
History
weeksweeks · through 2026-09
542
These move opposite each other about 48% of the time
Their swing sizes only faintly mirror each other (~14% of the pattern is shared).
Roughly random — these don't track each other in a meaningful way.
On a log scale so equal % moves take equal vertical space — best when one series has grown much faster than the other.
What to Watch
USD/HKD Exchange Rate moves ~11 weeks before Unemployment Rate
Watch USD/HKD Exchange Rate for an early read on Unemployment Rate.
Decouples in drawdowns
The relationship weakens when both prices are falling — don't count on this pair as a hedge under stress.
Flips between sync and inverse
Sometimes the two move together, sometimes opposite. Don't treat this as a stable signal.
Explore
Advanced
Statistics
In sync(i)
48.4%
Headline metric
Movement correlation(i)
-37%
Based on values
95% CI
-44% → -30%
Likely range of correlation
Pipeline
Pipeline Summary
542 paired data points survived the weekly window.
Raw input
938
11,370
Normalized
938
11,370
Prepared
938
2,365
Aligned
542
542
Invalid removed
R²(i)
13.7%
Variance explained
Significance
p < 0.001
Statistical confidence
Data points
542
Robust
Time-Shifted Correlation
See how correlation changes when one series is offset in time. A taller bar at a non-zero shift means the two move together better when one leads the other — that's a potential lead/lag signal.
Correlation by shift
Click a bar to inspect. Range: -12 to 12 weeks.
Selected shift
-11 weeks
Correlation at this shift
-45%
+8% stronger than no-shift baseline
USD/HKD Exchange Rate shifted 11 weeks later. Reads: "Does Unemployment Rate today line up with USD/HKD Exchange Rate 11 weeks ago?"
531 overlapping points at this shift
Baseline
-37%
No-shift correlation, matching the main time-series chart above.
Peak shift
-11 weeks
-45%
A non-zero peak suggests one series lines up better when shifted against the other.
Stability
How the correlation evolves over time. A stable line means the relationship is reliable; large swings signal regime-dependent behavior.
Do They Crash Together?
How these series behave when markets are rising, falling, or diverging. A correlation that holds in drawdowns is very different from one that only works in rallies.
Both Rising
+49%
47 periods · Return correlation when both series rose
Both Falling
-5%
56 periods · Return correlation when both series fell
Diverging
-12%
146 periods · Return correlation when series moved apart
Scatter
0
A: 0 / B: 0
Duplicates removed
0
A: 0 / B: 0
Alignment drops
2,219
A: 396 / B: 1823
Series A
Unemployment Rate
UNRATE
FRED · 938 raw → 938 prepared
Series B
USD/HKD Exchange Rate
DEXHKUS
FRED · 11,370 raw → 2,365 prepared
Sign agreement
100.0%
How often both values share the same sign.
Zero crossings
9
Estimated crossover points between normalized spreads.
Slope
-0.0877
Linear regression slope.
Intercept
8.2046
Linear regression intercept.
Saved 4 days ago · ID: fred-dexhkus_fred-unrate_5y
Explore
Top 10 by absolute correlation
Ranked across both sides of this comparison using the same dense row format as the single-symbol correlations view.