Using a Blattistics Regression for Risk Management

Blattistics regressions allow you to minimize risk, maximize risk, or place hedges. The following page is a tutorial in risk management based on a sample single variable regression shown below:







Reading the Model

The above regression shows the predicted stock prices of Wal-Mart (WMT), General Electric (GE), and Proctor & Gamble (PG) based on the price of the Dow Jones Industrial Average (^DJI). The Dow Jones Industrial Average is considered to be representative of the market at-large, and is assumed to have a substantial causal effect on most stocks.
The first thing to look at when viewing a Blattistics regression model is the T-Statistic (highlighted for GE in a black box). If the T-statistic is less than two in magnitude (-2< t <2), then the independent variable (^DJI) is a poor predictor of the dependent variable's share price. All of these have high magnitude T-Statistics, so this is not a concern here.

Elasticities*

The next value to consider is the elasticity. This is the likely rise in the dependent variable's share price (WMT, GE, and PG) given a 1% increase in the price of the dependent variable (^DJI). (Technical note: Elasticity is not the same as Beta from technical finance). GE's elasticity of 2.05 means that GE shares would likely increase by 2.05% if the Dow Jones increased by 1%. If you are confident that markets will go up then you might want to consider shifting some of your portfolio into stocks that are riskier than the market (elasticities greater than 1). PG's elasticity of 0.34 means that PG is likely to increase by 0.34% for a 1% increase in the Dow Jones. This is an example of a low-risk asset. You can reduce the risk in your portfolio by picking stocks with elasticities between 0 and 1. The third example, Wal-Mart, has a negative elasticity. This indicates that if the Dow Jones goes up 1%, WMT stock is likely to decline by 0.19%. Investing in stocks with negative elasticities is a good strategy if you are confident the market will decline.

Hedging

A hedge is an investment strategy that reduces price risk. All three of these stocks are likely to move depending on changes of the price of the Dow Jones Industrial Average. Consider a portfolio where the investor purchases $3 of Wal-Mart stock for every $2 of Proctor and Gamble stock purchased. When combined, these elasticities cancel out almost perfectly leaving our investor secure from market risk (the only surefire way to avoid market risk is to thoroughly diversify or not invest at all). If the investor identifies both stocks as undervalued (Click here to read more about using Blattistics regressions for value investing), a hedge investment could become very profitable. A fully diversified portfolio should have the weighted average of the price elasticities equal to 1.

Choosing an Independent Variable

This example considered the Dow Jones Industrial Average as the independent or causal variable. You might also want to consider the S&P 500 (^GSPC), the NASDAQ Composite (^IXIC), the ten year note (^TNX) if you believe your portfolio is affected by interest rate changes, the Nikkei average (^N225) if your portfolio is correlated to the Japanese markets etc. You might also want to consider choosing a leading company in an industry related to your portfolio. If you believe oil prices have a large effect on your portfolio consider choosing Exxon-Mobil (XOM). If you believe your portfolio is highly related to the high-tech economy you might want to choose Microsoft (MSFT). If you have large holdings in a single company, you should consider choosing that symbol as your independent variable. Just remember that many of the relationships picked up in these models may not be causal, and are instead purely coincidental. Thank you for your interest in Blattistics and good luck.

*Elasticities are not currently included with Blattistics multiple regressions.

Order a Single Regression
Previous Page.