Investor sentiment or information content? A simple test for investor sentiment proxies

Abstract

This study suggests a simple test to assess the validity of employing a variable as an investor sentiment proxy by examining whether the variable exerts a long-run impact on prices. We also conduct the test using news sentiment indices for the U.S. and Korea, which represent a developed and influential market and a leading emerging market, respectively, as potential proxy variables. The test results, on the one hand, lead to a pessimistic view of employing news sentiment as a proxy for investor sentiment. On the other hand, the results also demonstrate a connection between news sentiment and investor sentiment, particularly in Korea. We suggest that a variable may not be appropriate as an investor sentiment proxy even when the variable is somewhat related to sentiment.