Out-of-Sample Equity Premium Predictability in South Africa: Evidence from a Large Number of Predictors. Issue 8 (2nd August 2016)

• Record Type: Journal Article
• Authors: Gupta, Rangan; Modise, Mampho P.; Uwilingiye, Josine
• Is Part Of: Emerging markets finance & trade. Volume 52:Issue 8(2016)
• Page Start: 1935
• Page End: 1955
• Publication Date: 2016-08-02
• Subjects: bagging -- Bayesian predictive regressions -- equity premium -- forecast combinations -- predictive regressions -- principal component regressions; Balkan Peninsula -- Commerce -- Periodicals; Balkan Peninsula -- Foreign economic relations -- Periodicals; Europe, Central -- Commerce -- Periodicals; Europe, Central -- Foreign economic relations -- Periodicals; Europe, Eastern -- Commerce -- Periodicals; Europe, Eastern -- Foreign economic relations -- Periodicals; Turkey -- Commerce -- Periodicals; Turkey -- Foreign economic relations -- Periodicals; Asia -- Commerce -- Periodicals; Asia -- Foreign economic relations -- Periodicals; Eurasia -- Commerce -- Periodicals; Eurasia -- Foreign economic relations -- Periodicals; 382.05
• DOI: 10.1080/1540496X.2015.1058075
• Languages: English
• ISSNs: 1540-496X

Abstract

ABSTRACT: This article uses a predictive regression framework to examine the out-of-sample predictability of South Africa's equity premium, using a host of financial and macroeconomic variables. We employ various methods of forecast combination, bootstrap aggregation (bagging), diffusion index (principal component), and Bayesian regressions to allow for a simultaneous role of the variables under consideration, besides individual predictive regressions. We assess both the statistical and economic significance of the individual predictive regressions, combination methods, bagging, principal components, and Bayesian regressions. Our results show that forecast combination methods and principal component regressions improve the predictability of the equity premium relative to the benchmark autoregressive model of order one (AR[1]). However, the Bayesian predictive regressions are found to be the standout performers with the models outperforming the individual regressions, forecast combination methods, bagging and principal component regressions, both in terms of statistical (forecasting) and economic (utility) gains.