The Japanese stock market has been non-directional and unstable for more than ten years and fully reflects the ailing Japanese economy. Looking around the world, it is hard to find any traditional financial products profiting steadily under this situation. What is an investor to do? Pair-trading is a relative value investment strategy that seeks to minimise market risk and take advantage during such unstable times.
The term relative value covers a variety of low volatility trading strategies with the consistent theme of attempting to reduce market risk. In other words, the manager seeks to generate a profit regardless of which direction the markets are moving. All relative value strategies minimise market risk by taking offsetting long and short positions in related stocks, bonds and other types of securities.
AGS specialises in pair-trading which means taking equal and offsetting long and short positions in two different stocks which have diverged from their recent normal price patterns. From a statistical viewpoint, securities tend to form short-term price patterns based on the cumulative expression of market sentiment. But no investment strategy is without risk and all relative value investments strategies (including pair trading) seek to replace market risk with "correlation risk" where the prices for two securities are expected to follow a particular pattern.
How exactly does pair trading work?
If the market goes up, anyone can easily make profit by buying ETF or investing in a fund managed by a good manager. But when we are not sure the market will go up, can we not make a profit from the stock market? Is it impossible to get a profit in a bear market? The answer to both questions is "NO". Let's look at exactly how pair trading works in very simple terms by taking the example of two stocks A and B.
You are not confident of a recovery in the market and not sure if the Nikkei or Topix will go up or not but you believe stock A's business, products and market are attractive. Still, considering the market, you can't feel sure stock A will go up. At the same time, you think stock B is clearly worse than A and you feel certain the performance of stock B will not surpass the performance of stock A in any case. If your analysis is correct, you can make profit by doing a pair trade as shown.
Suppose you buy stock A with 1 million yen and sell short stock B with the same amount. Here are two possible outcomes.
Scenario 1: The Nikkei drops 5 % over a month
Stock A couldn't avoid falling but only by 2 %. Stock B reacted more to the down market and it dropped 6%. Your A investment sees a 2% (20,000 yen) loss and stock B sees a 6% (60,000 yen) profit. If you exit the pair you could net 40,000 yen.
Scenario 2 The Nikkei rises 5% over a month
Stock A (the 'good' stock) rises faster than the index reaching 7%. Stock B (the 'bad' stock) also goes up but only by 3%. You see a 70,000 yen profit on A and a 30,000 yen loss on B. If you exit the pair the net profit could be 40,000 yen.
How to identify pair-trade opportunities
Once you understand the basic pair-trade concept the question quickly becomes what stocks to buy and sell, in other words, how to match up the pairs. It is also critical to know when to enter the trades and when to exit. While there are many ways of matching pairs and identifying the timing for pair-trade opportunities, the three main ways to study pairs of securities are as follows:
(1) Fundamental approach
The fundamental approach is the relative comparison of fundamental data using indexes and other reports (see list). With this approach, comparisons between similar types of securities are generally going to be more effective. The holding period for these pairs tends to be longer than the other approaches. Timing for the entry and exit is more difficult to formalise because fundamental data is not real-time.
(2) Technical approach
This approach selects securities using one or more types of technical analyses. Merits of this approach are that securities can be selected across sectors and industries and the timing signals, being based on market data, can be readily tracked. This approach depends entirely on the effectiveness of the indicators used. Pros consider the following list of indicators to be reasonably effective.
(3) Statistical approach
This last approach works by applying advanced statistical models to historical market data using sophisticated computer programmes. This approach often includes aspects of both 1 and 2 above and, therefore, it can be said that it allows for the widest possible analysis. This approach requires powerful computers and advanced programming skills to implement. The concept of "mean reversion" is frequently at the heart of this approach. Mean reversion means that the relative prices of correlated securities tend to revert to a statistical average over a specified time frame.
AGS managers develop and operate highly advanced statistical modelling software and pair screening filters. We conduct seminars from time to time in Tokyo for those interested in learning more about our methods and practices. Currently, the clients for our technology include Japan's top regional banks and investment corporations and the clients for our investment partnerships include wealthy individuals seeking good returns with relatively low value at risk.