Can economic policy uncertainty, oil prices, and investor sentiment predict Islamic stock returns? A multi-scale perspective

Abstract

This study investigates the causal relationships between economic policy uncertainty, oil prices, investor sentiment, and stock returns of nine Dow Jones Islamic Market Indices. Specifically, this study analyzes the causal effect of these variables on Islamic stock returns at different time scales using the ensemble empirical mode decomposition model. First, we decompose the economic policy uncertainty proxy, oil prices, and investor sentiment into different independent components called intrinsic mode functions (IMFs): short-term IMFs designate the effects of irregular events; medium-term IMFs present the effects of extreme events; and long-term IMFs capture long-term effects. Second, we employ a nonlinear non-parametric causality model to test the causal relationship between different variables and Islamic stock returns at both the original and decomposed levels. We find causal relationships between the underlying variables and Islamic stock returns in several time frequencies rather than in the whole sample period. Our results suggest that the use of lagged economic policy uncertainty, oil prices, and investor sentiment may improve the predictability of Islamic stock returns. A test of forecast accuracy indicated the robustness of our results.