The Compass Rose Pattern of the Stock Market: How Does it Affect Parameter Estimates, Forecasts, and Statistical Tests?

A "compass rose" pattern sometimes appears when stock returns are plotted against themselves with a one-day lag, since stock prices move in discrete steps. In this paper, we perform a Monte Carlo study on simulated stock price series rounded in different ways to mirror the behavior of stocks on the Stockholm Stock Exchange. We find AR-GARCH parameter estimates to be affected by the discreteness imposed by rounding. Based on the compass rose and the discreteness, we investigate, theoretically and empirically, different possibilities of improving predictions of stock returns. The distributions of the BDS test as well as Savit and Green's dependability index are also influenced by the compass rose pattern. However, throughout the paper, we must impose unrealistically heavy rounding of the stock prices to find significant effects on our estimates, forecasts, and statistical tests.