CAPM / Quant. help......

So I am a level I candidate for December… I have reviewed the capital asset pricing model, thought I had a pretty firm grasp of it.

Anyways, I am still an undergrad… I am having trouble with a case assigned by my professor. I am following the attached excel file as a guide for my own project. I am using a different case / different scenario. This file is just an example of the format to be taken…

webpage.pace.edu/pviswanath/class/data/ betacomp. xls

My question is in the regression tab. They run a regression of the stock “ALCO” vs the market returns. How do they calculate the Predicted Y and Residuals for the 60 observations at the bottom? Are the ‘residuals’ the alpha’s of the stocks?

I’m not sure whether this is the appropriate place for this discussion, but I’ll give you some pointers nonetheless.

The predicted y-s are the fitted values. You get those by plugging the x-values in the estimated least squares model. The residuals are the actual returns less the fitted values (ie. how much the model is off). Notice that this is a regression of ALCO returns vs. market returns for 60 periods. Hence, it is a security characteristic line for the stock. The slope corresponds to ALCO’s beta, while the intercept is its alpha.

Also, before jumping to any conclusions, be mindful of the significance of each estimate. For example, yes the alpha is negative, but we can’t reject the null that it is 0 at any reasonable level of significance.

Good luck. Study more.