Empirical distributions of stock returns: scandinavian securities markets, 1990-95

Abstract

The assumption that stock rctums are normally distributed has long been disputed by the data. In this article we test (and clearly reject) the normality assumption using time series of stock retums for each of the four Scandinavian markets during the first half of this decade. More importantly, we fit to the data four altematiw specifications, find empirical support for the scaled-t distribution, and quantify the magnitude of the error that stems from predicting stock retums by lIsing a Nom1al distribution. Ke)