The returns achieved by investors using quantitative investment disciplines such as tactical asset allocation depend heavily on the quality of the statistical models employed. Investors and investment professionals must be able to assure themselves that the statistical work they rely upon is of the highest quality. The question is, what tests can give this assurance?

A standard of appropriateness calls for tests that can indicate whether statistical methods and the data to which they are applied are appropriate to each other. Such a standard needs to be applied before modeling work is undertaken and its results promoted.

Hinich has recently developed tests to determine when familiar linear techniques, including the majority of statistical regression methods, may be used. Applications of Hinich tests to stock and bond return data show that linear methods are not always appropriate. Only when the right tests are used can investors expect a confluence of data and method that will yield usable results.