Partner Up: How Pairs Trading Works with Market Relationships and Correlations – Ticker Tape
When a relationship between a pair of stocks, futures, or options gets out of line, pairs trading may offer potential opportunities. Here’s how some investors choose among many combinations of bullish and bearish positions.
Photo by Getty Images
Table of Contents
- Before applying pairs trading strategies, make sure you understand financial market relationships
- Know how closely a pair of stocks, indices, or futures are correlated
- When normally strong correlations part ways, it could potentially signal a pairs trading opportunity
Have you ever found yourself in the middle of a “typical” market situation that suddenly didn’t feel so typical? If so, you might consider exploring one of the more sophisticated market strategies: pairs trading.
So, what is pairs trading? It involves the simultaneous trade of two correlated securities. Some securities tend to move in the same direction, even if the percentage changes are different. But sometimes the correlation between the two related securities gets out of whack, especially to an extreme level. That could signal a potential opportunity for a pairs trade, which aims to take advantage of the (presumably temporary) gap.
Getting Cozy with Pairs Trading
Relationships and correlations exist throughout the financial markets. One of the more obvious relationships is between the S&P 500 Index (SPX) and the Dow Jones Industrial Average ($DJI)—they generally move in the same direction. But some relationships may not be so apparent. Take corn and crude oil futures. Why might they be related? Because more than 40% of harvested corn is pegged for ethanol production. Meanwhile, corn futures and soybean meal futures compete as substitutes to nourish cattle. Then there’s gold versus silver, oil versus natural gas, bonds versus stocks … and on it goes.
There are a few essential concepts to really understand pairs trading. Let’s explore.
Correlation Between Pairs
Correlation has to do with how a pair moves in the same or different directions at the same time. Statistically, correlation is measured from -1.00 (where one asset goes up and the other tends to go down) to +1.00 (highly correlated pairs where the two are expected to move in the same direction).
When you look at two stocks or indices as a potential pair, a positive 0.60 or higher correlation is a good place to start. If the correlation is lower than that, the relationship between the two stocks may be weak. And when the long stock in the pair goes down and the short stock either goes up or doesn’t drop, it can lead to surprises.
To find pairs trading correlation values, go to the Charts tab on the thinkorswim® platform from TD Ameritrade:
- Type SPX–NDX in the symbol field to pull up a sample pairs chart (see figure 1). If the pair is moving up, it means SPX is outperforming NDX. But if the pair is moving lower, as it is in figure 1, NDX is outperforming SPX.
- Select Studies > Add study > All Studies > O–P > PairCorrelation. Now you can see the correlation between the two symbols you entered.
Enter any two symbols in the same way (e.g., MNKY–XYZ) to find their correlation.
You can use any two correlated assets for a pairs trade. For example, you can use stocks as long as the two listings are correlated and have some type of relationship. But a long stock/short stock pairs trade can use up a lot of trading capital and comes with significant risk if the pair should move against you. Options, particularly options spreads like verticals, can offer lower capital requirements and defined risk. For example, you can allocate a relatively small amount of capital to a bullish vertical in one symbol and a bearish vertical in another. Because the risk is defined, even if the pair’s correlation collapses (your bullish stock crashes and/or your bearish stock rallies), the verticals won’t exceed a specific max potential loss. Keep in mind options still have greater and more complex risk than stocks and options trading is not for everyone.
Notional Value for Pairs
It’s also important to be aware of the notional value when putting on a pairs trade. Suppose you buy 100 shares of stock A for $100 and short 100 shares of stock B for $50 as a pairs trading strategy. The two stocks are highly correlated, and you think stock A will outperform stock B by rising more (or dropping less).
If both stocks drop, but stock A drops 5% to $95 and stock B drops 6% to $47, stock A has performed better than stock B because it dropped less. But how does your profit and loss work out? The long shares of stock A have lost $500. The short shares of stock B have made $300. You were right on your assessment of the pairs trade, but it still lost money. Why?
You incurred a loss because the notional value of the long shares of stock A was 100 x $100 = $10,000. The notional value of the short shares of stock B was 100 x $50 = $5,000. Stock A had twice the notional value of stock B, so the pairs trade was weighted heavily toward stock A.
The difference in notional values might not be a big deal for two stocks that have roughly the same price. But it becomes a bigger deal if you pairs trade with futures or index products.
Suppose you think oil prices are going to decline and natural gas prices will rise. You might establish a pairs trade by selling light sweet crude oil futures (/CL) and buying natural gas futures (/NG). Multiply the value of a one-point move in the underlying by the prevailing price. If /CL is trading at $42 and /NG at $2, here’s what it would look like:
- /CL notional value = $42 x 1,000 = $42,000
- /NG notional value = $2 x 10,000 = $20,000
A /CL contract is more than twice the size of an /NG contract, so if you’re considering a pairs trade, you might think about buying two /NG contracts for every /CL contract sold. But that notional value is pretty rich.
Keep in mind that futures trades have many moving parts. You’ll need confidence, experience, and a large account. But the concepts discussed here can be applied to any pairs trade.
Different Types of Pairs Trading Strategies
An intra-sector pairs trade can involve two stocks in a particular industry like technology, health care, or energy, where correlations are often high. Let’s consider two hypothetical large oil companies, MNKY and XYZ. Suppose you chart the “price” of the pair MNKY–XYZ. If you believe the pair price will continue to go higher, and MNKY will continue to outperform XYZ, you could buy shares of MNKY and sell shares of XYZ. But if you believe the pair price will go lower, you could sell MNKY and buy XYZ.
Next, we’ll look at an inter-index pairs trade. Suppose technology stocks are booming and the NASDAQ Composite (COMP) is outperforming the SPX. You could hypothesize that eventually the Technology sector will come back down to earth and COMP will resume its normal, in-line correlation with SPX. In such a scenario, a pairs trading strategy example could be to go long SPX and short COMP via exchange-traded funds (ETFs), options, or futures. You’re looking for the correlation to come back in line, resulting in a potential profitable trade.
You could also consider an intermarket pairs trade. Suppose lower rates have pushed Treasuries and U.S. stocks higher, at least in the short term. You think Treasuries have reached a top but that SPX still has room to move higher. You could consider a pairs trade that’s short Treasury futures and long SPX futures.
Very Important: Keep Pairs Trade Durations Short
Pairs trading strategies are usually meant to be short to medium term at the longest. Even the most highly correlated names tend to lose correlation over time. Pairs trades can be complex, and it’s important to monitor your risk closely. What if the Technology sector continues to outperform the broader market? Or what if the pair’s divergence from its normal correlation lasts longer than you expected? Either scenario could result in a losing trade.
The idea behind a pairs trade is to take advantage of divergences in highly correlated securities. To increase the odds of success, track relationships over time, identify price divergences, and hypothesize when they may come back in line.