Alternative methods of computing estimates of inequality measures from grouped data are critically examined in terms of their theoretical and empirical properties. The use of a simple “split-histogram” technique of interpolation is explained and supported. Theoretical and empirical support is also provided for the “⅓/⅔ rule”—a simple computational procedure for a point estimate of an inequality measure derived from its standard grouping bounds.

It is often desirable to know whether or not a risky asset's beta coefficient has changed and, if so, at what point in time the change occurred. For example, this knowledge is of obvious importance to beta-using security analysts and portfolio managers. As another example, a given theory may imply that a particular firm's beta should have changed at different points in time. Investigators may want to test such a hypothesis. Furthermore, tests are frequently performed on the effects of events on residuals of the market model, tests requiring the assumption of beta stability. For these, and possibly other reasons, it is useful to be able to detect that a change in beta did, in fact, take place as well as, in some instances, identifying the point in time at which the change took place.