Portfolio Regression Help!?

Question

The CFO of Hankcock, Al Dipta, is trying to estimate the cost of equity of his company using the CAPM, for which he needs an estimate of the beta of the company's stock. He could use the estimate published in Value Line, but he would prefer to come up with an estimate of his own, as he feels he could get a more accurate estimate of the beta of a small airline such as his by regressing the company's historical returns on the returns on a select group of transportation stocks rather on a broad portfolio of stocks as done by Value Line analysts? Is this a good thing to do? why or why not?

Can someone help me to know the basics of the question?

Answer 1

One of the ideas behind using beta is that there are two kinds of risk -- market risk and diversifiable risk. We get rewarded for taking on market risk and do not get rewarded for taking on diversifiable risk. Beta just measures the market risk.

His plan is to calculate expected return based on the returns of the sector. While the sector has less diversifiable risk than an individual company -- it still has diversifiable risk. Therefore, doing this regression is basing the expected return of a stock on an index that includes risk factors that we are not rewarded for taking on.

The second problem with this idea is that beta involves regressing the EXCESS returns of the asset against the EXCESS returns of the market (meaning we look at the raw return and subtract off the return of the risk free asset). Using the method outlined above doesn't incorporate changing returns for the risk free security.