Statistical peers
Statistical peers
In plain words
Every stock has “neighbors”: stocks whose price almost always moves like its own. When these neighbors move clearly together and the stock has not followed yet, it tends to catch up. The strategy takes a position in the direction of that catch-up.
A statistical peers signal, number by number
This is what a statistical peers signal looks like in PairScanner. The stock and the figures are fictional.
Click a number to learn its role
Expected move+2.7%over 20 sessions, versus market and sector, after costs
My role: tell you what this kind of signal earned on average in the past, once costs are paid. Here +2.7% over 20 sessions, on top of what the market and the sector did.
80% range-3.0% to +8.2%median +2.2%
My role: show the range of the most common outcomes. Eight similar signals out of ten ended inside it. The narrower it is, the more predictable the outcome.
Probability62%of a positive outcome
My role: estimate the chance that this signal ends positive, after costs. 50% would be a coin flip; 62% means about six similar signals out of ten ended positive.
Sample402 / 650positive cases, 2019 to 2026
My role: tell you how many past situations the figures rest on. 650 cases make a solid base: the figures do not hang on a few lucky trades.
Worst decile-4.8%one outcome in ten did worse
My role: give the size of an unfavorable outcome, so you can size the position knowingly.
ValiditySep 29, 2026expires Oct 27, 2026
My role: date the data used and tell you how long the signal counts. An expired signal is no longer tracked.
What PairScanner does
Every month, the algorithm finds for each stock the three stocks whose returns were closest to its own over the past year, 252 sessions. These are its statistical peers, chosen by the data rather than by a sector classification.
Every evening, it computes the peers' average move over the last five sessions, removes what the market, the sector and the stock's recent past explain, and checks whether what is left is unusual. Beyond two standard deviations, a signal appears in the peers' direction.
It then computes, for 5 and 20 session horizons, the expected move after costs, the estimated probability of a positive outcome, the range of plausible results and the size of an unfavorable outcome, from replaying the same rule on history.
You code nothing and calculate nothing: you open the Statistical peers tab of the dashboard and read today's result.
The idea in detail
Two companies can look alike on the stock market without sharing an official sector: same customers, same sensitivity to rates, same supply chain. Rather than guessing these links, the strategy measures them: peers are the stocks whose prices moved most closely together.
When peers make an unusual move together, they often signal information that also concerns the stock. If it has not reacted yet, the gap tends to close through a catch-up.
It is the intuition of pair trading, applied to a group of peers rather than a single twin.
A worked example
Three chipmakers are the statistical peers of a fourth company. Over five sessions, they gain 6% on average, while the market and the sector only explain a 2% rise.
| Step | Value |
|---|---|
| Peer 1 | +7% |
| Peer 2 | +6% |
| Peer 3 | +5% |
| Peers' average | +6% |
| Part explained by market and sector | +2% |
| Unusual move of the peers | +4%, four times the usual |
| Target stock over the same period | 0% |
| Decision | Buy |
The fourth company has not followed its peers yet: the strategy bets on its catch-up.
Fictional example for teaching purposes. It is neither a signal nor a recommendation.
Why it would work
In 2019, Huafeng Chen, Shaojun Chen, Zhuo Chen and Feng Li tested a very similar strategy on US stocks: for each stock, a group of stocks with the most correlated past returns. Stocks that had diverged from their group tended on average to catch up, with large abnormal returns. The authors link most of these gains to two effects PairScanner also computes: short-term reversal and sector momentum.
The explanation lies in investors' limited attention: news affecting a whole group of companies is not built into every price at the same moment.
What research says today
In 2019, Charles Lee, Stephen Teng Sun, Rongfei Wang and Ran Zhang showed that returns of technologically close companies help predict those of a given company: links invisible in official sector classifications carry information. A long-short strategy built on these links earned about 1.2% per month in risk-adjusted returns in their study.
Lauren Cohen and Andrea Frazzini found the same in 2008 for customer and supplier links, the starting point of the lead-lag strategy.
Sources
Studies cited on this page:
- Huafeng Chen, Shaojun Chen, Zhuo Chen and Feng Li (2019). Empirical Investigation of an Equity Pairs Trading Strategy. Management Science, 65(1), 370-389.
- Charles M. C. Lee, Stephen Teng Sun, Rongfei Wang and Ran Zhang (2019). Technological Links and Predictable Returns. Journal of Financial Economics, 132(3), 76-96.
- Lauren Cohen and Andrea Frazzini (2008). Economic Links and Predictable Returns. The Journal of Finance, 63(4), 1977-2011.
PairScanner promises no results: the figures describe what the strategies did in the past. Educational content, not investment advice; PairScanner places no orders.
See today's statistical peers signals