Using a Blattistics Regression to Identify Value Opportunities

Blattistics regressions can give you an estimate of whether a stock is overpriced or underpriced. The following is a tutorial on identifying value opportunities based on a sample multiple variable regression shown below:






Is this a good model?

Before using a Blattistics regression to valuate a stock, it is necessary to check to see if the model shows a strong or weak relation between the variables. The above regression shows the predicted impact of the S&P 500 (^gspc) and the ten year note (^tnx) on BP and Microsoft (MSFT) respectively. An R2 value close to one is a good indicator of a highly predictive model. The T-Statistics show whether the regression coefficients are statistically significant. If the T-statistic is less than two in magnitude (-2< t <2), then there is a high likelihood that the regression variable (^DJI) is a poor predictor of the dependent variable's share price. As you can see above, the first model is statistically more sound than the second model, although both seem to show a close linear relation.

Fair Value

Blattistics regressions can be used to estimate fair value and identify "value investing" opportunities. Value investing is the strategy of finding stocks which are undervalued. This has been popularized by the saying "Buy low and sell high." The current price of the stocks (BP and MSFT) are shown in the black circles above, and the estimated fair value of the stocks are shown in the red circles above. In the example, BP is slightly overpriced and Microsoft is slightly underpriced. Please note: when you purchase a Blattistics regression a low fair value is likely the result of a recent upsurge in the price of the stock, and a high fair value is likely representative of a recent decline in the price of the stock relative to the independent (X) variables. However if you wait several weeks or months and calculate the fair value of the stock using the coefficients in the model, you are likely to see a greater variance between the share price and fair value. If fair value is substantially below the share price, you might want to consider selling the stock. If fair value is substantially above the share price, you might want to consider buying the stock. While other considerations such as earnings and corporate news may cause a stock to deviate from its fair value, nonetheless this method can be useful in identifying value opportunities, particularly among smaller companies which are not watched as closely and are subject to greater cyclical change.

Blattistics also sells Adjusted P/E Spreadsheets which are intended to provide a searching tool to investors looking for value stocks. Click here to read a tutorial on using Adjusted P/E Spreadsheets to find value stocks.

Order a Single Regression
Order a Multiple Regression
Previous Page.