🤖 Quick Answer

What is the PairTradingPRO Correlation Screener?

It is a web-based tool for finding asset pairs for pair trading and statistical arbitrage across Crypto, Forex, and U.S. Stocks. For each pair, the screener calculates:

• ✅ current Pearson correlation;
• ✅ rolling statistics (average, minimum, maximum, stability);
• ✅ correlation over 30-, 90-, and 180-day windows;
• ✅ maximum correlation drawdown;
• ✅ R²;
• ✅ Spearman correlation;
• ✅ the gap between Pearson and Spearman;
• ✅ a final Health Score from 0 to 100.

Selected pairs can go straight to spread building and automated trading in the PairTradingPRO desktop terminal. The web version is free for 7 days.

---

🎯 Why You Need a Correlation Screener

Finding a good pair by hand among hundreds of instruments is nearly impossible. You would have to download price data, calculate correlations over several windows, check for outliers and stability, and then repeat all of it every day.

The Correlation Screener does this automatically and answers the three key questions every pair trader asks:

1. 🔗 Is there a relationship? How closely do the assets move together?
2. ⏳ Is it stable? Or is it a coincidence from the last few weeks?
3. 🧼 Is it clean? Or do a few abnormal candles create the high correlation?

---

⚙️ Calculation Settings: The Top Panel

Above the table you can see the conditions behind every number:

• ⏱ Timeframe 1h — calculations use hourly candles.
• 📅 Window 360d — the main analysis window: 360 days of history.
• 🔄 Updated — the date and time of the last recalculation. Always check that you are looking at fresh data.

🎛 Filters and Sorting

 

 

---

🧩 Every Screener Column Explained

The table has five logical blocks. Let's go through them from left to right, using real examples from the screener.

1️⃣ COUPLE: Pair and Market

Couple — the two assets whose relationship the screener tests. The tool icon to the left of the name opens the pair for further work.

Category — the market the pair belongs to:

• 🪙 Crypto — cryptocurrencies (PAXGUSDT – XAUTUSDT, BTCUSDT – ETHUSDT);
• 💱 Forex — currency pairs (HKDJPY – USDJPY, EURUSD – USDDKK);
• 📈 Stocks — U.S. equities (NSC.US – UNP.US, DAL.US – UAL.US, MA.US – V.US).

💡 Pro tip: the best pairs almost always have an economic reason behind them. PAXG and XAUT are both gold-backed tokens. The Hong Kong dollar is pegged to the U.S. dollar, so HKDJPY follows USDJPY. The Danish krone is pegged to the euro, which explains the strong inverse link in EURUSD – USDDKK. NSC and UNP are railroads, DAL and UAL are airlines, MA and V are payment networks, and AMAT and LRCX make chip-making equipment. If a high correlation has no fundamental reason, be cautious.

---

2️⃣ CURRENT CORRELATION

Current ρ — the Pearson correlation coefficient at this moment, ranging from −1 to +1:

• ⬆️ close to +1 — the assets move in the same direction (PAXGUSDT – XAUTUSDT: 0.9942);
• ⬇️ close to −1 — the assets move in opposite directions (EURUSD – USDDKK: −0.9735);
• ⚪ close to 0 — no relationship.

Both extremes work for pair trading. With positive correlation, the spread is built as "long one / short the other." With negative correlation, both legs are opened in the same direction.

|ρ| Bar — the absolute correlation as a percentage, with a progress bar. The sign is ignored and only the strength is shown: −0.9735 becomes 97.4%.

Power — a text label for the relationship strength. In the screener, pairs with |ρ| of 0.90 or higher are labeled Very strong (DAL – UAL at 0.9021). Pairs just below 0.90 are already Strong (AMAT – LRCX at 0.8997, BTC – ETH at 0.8872).

---

3️⃣ ROLLING STATISTICS

Current correlation is a single snapshot. Rolling statistics show how the relationship behaved across the whole 360-day window.

Average ρ — the average correlation over the period. If it is close to the current value, the relationship is behaving normally. If the current value is well above the average, it may be a temporary spike.

Min ρ / Max ρ — the range the correlation stayed in. In the interface, the minimum is labeled "Mines ρ." The narrower the range, the more reliable the pair:

• 🟢 PAXGUSDT – XAUTUSDT: from 0.9915 to 0.9964 — a very tight range;
• 🟡 BTCUSDT – ETHUSDT: from 0.8401 to 0.9178 — strong, but it moves around noticeably more.

⚠️ Note: for negatively correlated pairs, read the minimum and maximum the other way around. For EURUSD – USDDKK, the "minimum" of −0.9824 is the strongest point of the relationship, and the "maximum" of −0.9581 is the weakest.

Stability — the overall verdict on how steadily the relationship holds. Stable means the correlation doesn't jump around and the pair can be considered for trading.

---

4️⃣ MULTI-WINDOW

This block checks whether the relationship looks the same over different horizons and how good it is.

30d / 90d / 180d — correlation over the last 30, 90, and 180 days. It is an early warning that the relationship is starting to break down:

• ✅ consistent values (PAXG – XAUT: 0.993 / 0.994 / 0.994) — the relationship holds on every horizon;
• 🚩 30d well below 180d — the relationship has been weakening recently, which is a red flag;
• ➖ a dash "—" (for some stocks on the 90d and 180d windows) — no value was calculated for that window. Treat these pairs with more caution.

Max DD (maximum correlation drawdown) — the largest drop of rolling correlation from its peak during the period. ❗ This is the drawdown of the correlation itself, not of price or spread. The lower the value, the less often the relationship "broke":

• PAXG – XAUT: 0.0049 — the relationship barely weakened;
• BTC – ETH: 0.0846 — there were periods of noticeable weakening.

R² (coefficient of determination) — the share of one asset's movement explained by the other. Here R² equals the square of the current correlation: 0.9942² ≈ 0.9885. So about 99% of PAXGUSDT's moves are described by XAUTUSDT. For BTC – ETH it is about 79%: a good relationship, but each coin still has a sizable share of its own movement.

Spearman — rank correlation. Pearson measures linear relationship and reacts strongly to sharp outliers. Spearman checks whether the assets move in the same direction and barely reacts to individual abnormal candles. It is the main way to check whether the Pearson correlation is honest.

---

5️⃣ PAIRTRADINGPRO METRICS

|Δ P-S| (Pearson–Spearman gap) — the absolute difference between Current ρ and Spearman. For example, |0.9942 − 0.9783| = 0.0159.

• 🟢 Small value (up to ~0.03) — both metrics agree; the relationship is clean and even.
• 🔴 Large value — the high Pearson correlation is probably driven by a few sharp moves rather than a steady everyday relationship. For example, DAL – UAL shows 0.0780, well above the leaders in the list.

Health (Health Score) 🏆 — the overall pair score from 0 to 100. It combines strength, stability, consistency across windows, and cleanliness of the relationship into one number. The table is sorted by it by default.

 

---

⚠️ Important: A High Health Score Doesn't Mean Maximum Profit

This is a common beginner mistake. Pairs with a Health Score of 97–99 are usually linked rigidly, through a currency peg or a shared underlying asset. Their spread therefore moves in a narrow range, and divergences are rare and small. After fees, there may be almost nothing left.

Also, correlation is not the same as cointegration. Correlation shows that assets move together. Pair trading cares more about whether the spread between them returns to its mean. So the screener is the first filtering step, not a ready trading signal.

---

🛠 How to Use the Correlation Screener: Step by Step

1. 🎛 Set the filters: choose your market, Stability = Stable, health ≥ 85, Min |ρ| ≥ 0.8.
2. 📅 Check consistency: 30d / 90d / 180d values should be close, with no dip on the short window.
3. 🧼 Check cleanliness: |Δ P-S| should be small and Max DD low.
4. 🧠 Find the economic reason: same sector, shared underlying asset, or currency peg.
5. 📐 Build the spread in Spread Builder and see how widely it moves and how quickly it returns to the mean. To view relationships across many assets at once, use the Correlation Matrix. For trade ideas, try the PairTrade AI Finder.
6. 🤖 Automate: set the pair and your entry and exit conditions in the PairTradingPRO desktop terminal. The robot watches the market 24/7 and opens both legs at the same time.

🔀 For cross-exchange arbitrage of the same asset, the ecosystem has a separate tool, CryptoArb Scanner. It accounts for the constant offset between exchanges and filters out false spreads.

---

🏁 Conclusion

The PairTradingPRO Correlation Screener puts everything you need to select pairs for statistical arbitrage into one table:

• 💪 relationship strength;
• ⏳ stability over time;
• 📅 consistency across windows;
• 🧼 data cleanliness.

The Health Score turns all these checks into a single number. The full ecosystem takes you from idea to automated trading: Correlation Screener → Spread Builder → Desktop Terminal.

🚀 Try the Correlation Screener free for 7 days and find your first pair for a market-neutral strategy.