Testing for Predictability in Panels of Small Time Series Dimensions with an Application to Chinese Stock returns

The few panel data tests for predictability of returns that exist are based on the prerequisite that both the number of time series observations, T, and the number of crosssection units, N, are large. As a result, these tests are impossible for stock markets where lengthy time series data are scarce. In response to this, the current paper develops a new test for predictability in panels where T ≥ 2 but N is large, which seems like a much more realistic assumption when using firm-level data. As an illustration, we consider the Chinese stock market, for which data is only available for 17 years but where the number firms is relatively large, 160.