Closed End Fund Discounts as Sentiment Indices

Abstract

This abstract was created post-production by the JFI Editorial Board.

This paper presents the results of a simple test of the hypothesis that the discounts and premiums on closed end funds are related to investor sentiment. While the results are encouraging they are not definite. The index used to measure sentiment is constructed from weekly survey data. It may well be that this data is too noisy to pick up all the changes reflected in the discounts and premiums. Alternatively it could be that the index itself is not the appropriate one and that a different construction, e.g. a lagged index or a moving average, might capture more of the changes. A second possible problem is that the sentiment index may merely reflect current market conditions rather than expectations. To test this a second variable consisting of the changes in a market index was added to the regression to see if the impact of the sentiment index is eliminated. The results are again supportive but not definitive. Considering the somewhat crude sentiment index used and the simple nature of the tests the results are encouraging. The hypothesized relationship was strongly significant in most of the regressions. Unfortunately, they are not definitive and further research is necessary to either refine the test or determine why the non-significant relationships could not be explained by the test here.