Alpha, Beta, and R-Squared: Investing Definitions Worth Knowing

Whether you're investing in individual stocks or baskets of them via mutual funds and exchange traded funds (ETFs), it's important to understand the terminology used in the investing world. This knowledge can help you decipher the jargon you're likely to encounter from investment-related news in newspapers, blogs, and on TV.

Two common terms that I see used a lot are Alpha and Beta. Their definitions are below.

A mathematical estimate of the amount of return expected from an investment's inherent values. It measures the difference between a stock's actual performance and the performance anticipated in light of the stock's risk and the behavior of the market. Alpha measure's a stock's risk adjusted performance. That is, it measures a stock's performance taking into account its beta, or sensitivity to the index, and the risk-free rate of return of a three-month Treasury Bill. For example, if a stock has a beta of 1.5, it would be expected to gain 15% when the index gains 10%. If however, the actually gains 20%, this excess return represents the stock's alpha.

A relative measure of the historical sensitivity of the stock's price to overall fluctuations in the New Your Stock Exchange Composite Index. A Beta of 1.5 indicates a stock tends to rise (or fall) 50% more than this index. The "Beta coefficient" is derived from a regression analysis of the relationship between weekly percentage changes in the price of the stock and weekly percentage changes in the NYSE Index over a period of five years. In the case of shorter price histories, a smaller time period is used, but two years is the minimum.

A measure of the squared correlation between a stock's performance and that of the broader market as measured by an appropriate index. In other words, it measures how reliable the stock's beta is in judging its market sensitivity. Though a little more esoteric, R-Squared is similar to Beta, but in this case tells you what proportion of a stock's risk is market-related, a figure that cannot be adjusted by diversification the way beta can. A completely diversified portfolio would be perfectly correlated to the market, indicative of an R-Squared figure of 1.0. An R-Squared of 0, on the other hand, indicates that the beta measurement is irrelevant to its actual performance.

1 Star2 Stars3 Stars4 Stars5 Stars (2 votes, average: 2.50 out of 5)


  1. R-Squared ranges from 0 to 100 NOT 0 to 1.

    • Darren Berardi

      Perhaps in some contexts, but it is commonly represented as a value between 0 and 1. Perhaps you're thinking of 0 and 100% which when expressed as a non-percentage is the equivalent of 0 to 1 e.g. .1 is the equivalent of 10%.

  2. A Morningstar data sheet tells me that a certain fund has a R-Squared of seven. If zero means perfectly uncorrelated and one means perfectly correlated to an index (S&P in the case of Morningstar), how can it be greater than one?

Leave a Reply

Your email address will not be published. Required fields are marked *