Time-Period Basis

Time-period basis refers to a problem with using time-series data in which the statistical result can change, depending on to the period of the sample data. This problem discourages using a short time series, as chance events are more likely to be reflected in the results.

In testing various investment strategies, the time-series basis can be significant. Idiosyncratic events or using a time frame that captures few business cycles can make strategies appear more successful than they really are. To minimize the time-period basis problem, one should use long periods of sample data.

Consider a stock market strategy focused on value stocks. A time-series study that compared value stocks with growth stocks, but only over the last decade or so, might conclude that growth investing outperforms value investing. In reality, however, it is a well-established fact, based on numerous time-series studies covering much longer periods, that in the long run, value stocks outperform growth stocks.