Online Program Home
  My Program

Abstract Details

Activity Number: 621 - Portfolio Choice, Stock Returns, Bankrupcty, and Default
Type: Contributed
Date/Time: Thursday, August 3, 2017 : 8:30 AM to 10:20 AM
Sponsor: Business and Economic Statistics Section
Abstract #322572
Title: Valid Tests of Stock Return Predictability
Author(s): Yang Guangyi* and Wang Mingjin
Companies: Guanghua School of Management, Peking University and Guanghua School of Management, Peking University
Keywords: Return predictability ; Bonferroni ; Grid bootstrap

When the degree of persistence of the lagged predictor is uncertain and the regression disturbance is correlated with the regressor's innovation, conventional tests of predictability can be invalid. Campbell and Yogo (2006) proposed a Bonferroni-type test, which gains more power. However, as Phillips (2014) pointed out, DF-GLS interval can be invalid in the region of stationary, which causes Campbell and Yogo (2006)'s test do not work well when the potential predictor is stationary. Following Cochrane (2008)'s argument, we assume that potential predictors, such as dividend price ratio, cannot be random walk. With this assumption, by combining Hansen (1999)'s grid bootstrap method which is asymptotically valid uniformly for predictor over local to unity and stationary region and Campbell and Yogo (2006)'s Q test, we propose a more valid Bonferroni-type test for return predictability. The simulation result in our work suggests that our test for predictability is uniformly valid for predictor over local to unity and stationary region.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2017 program

Copyright © American Statistical Association