Westerlund and Narayan predictability test: Step-by-step approach using COVID-19 and oil price data

In this note, we provide a step-by-step approach of Westerlund and Narayan (WN, 2012, 2015) predictability test using COVID-19 and oil price data. This is an important exercise because the WN model addresses three salient features of time series data, namely persistency, endogeneity and heteroskedasticity. We consider COVID-19 and oil price data as predictors of stock market returns for four Asian countries to demonstrate the applicability of the WN (2012, 2015) predictability approach. • This note demonstrates a step-by-step approach of the WN (2012, 2015) predictability test. • WN model accommodates three salient features of time-series data, namely persistency, endogeneity, and heteroskedasticity. • COVID-19 and oil price does not significantly predict stock returns of Japan, Russia, and Singapore (except in the case of South Korea).