 
Summary: Text Probabilistic Modeling in Molecular Evolution
Spring Semester 2011
Maria Anisimova & Ziheng Yang
Author: Elke Schaper. For comments please mail to eschaper@inf.ethz.ch
Homework 15.04.2011: Hypothesis testing primer
0.1 True or false?
a) false The significance level is value of the test statistic for which we
decide to discard the H0. In general, is not possible to assign a probability
of being true to a hypothesis, if not the whole set of possible hypotheses is
known.
b) false Decreasing the significance level means decreasing the range of
pvalues for which we reject H0. We will less easily reject H0, and the prob
ability of keeping H0 although it is not true increases. Hence, the power of
the test is rather expected to increase.
c) false Again, it is not possible to assign a probability of being true to a
hypothesis, if not the whole set of possible hypotheses is known.
d) false The probability that H0 is correctly rejected equals to the power
of the test.
e) insufficient A type I error occurs when the test statistic falls in the
rejection region of the test, given that the H0 is true.
