November 11th, 2010, 2:46 pm
Hey Hansi,I have got no particular test in mind since I don't have a data set. I am writing on a piece of work which deals with equivalence testing. That means that against common practice the equivalence is not in the null hypothesis but in the alternative. Which is from the statistical point of view the more correct way to establish equivalence of two parameters because only the rejection is a statistical proof. So I am looking for tests in the finance world on which I can apply my meta framework of equivalence testing. To make my work at least a little relevant for practical problems I wanted to check first which tests are commonly used in finance. Let me give you some suggestions of mine:- (Augmented) Dickey?Fuller test (check for random walk, probably won't get rejected if H_1: random walk)- t-test, F-test, Durbin-Watson test, Breusch?Pagan test, White test (check regression model of CAPM)- Jarque?Bera test (check normality assumption, probably won't get rejected if H_1: normality)My practical experience is more of the "front office - don't bother me with too much math - style", so I can't really say, whether anyone uses these tests. Can you? Any more suggestions regarding other tests of interest (parametric, two-sided)?Thanks for the help!Steilermeier