23 Flashcards
Hypothesis?
A statement about a population parameter (theta). It can stand for mean, proportions, variance, quantile, etc.
Simple hypothesis?
Statement about population parameter (theta) that completely specifies the parameter. It says exactly what you think it is.
Compound hypothesis?
Statement about parameter (theta) that does not completely specify the parameter.
Hypothesis test?
Procedure to decide (chance that the procedure decides incorrectly) based on a sample which two complementary hypothesis is true. One is H0 while other is HA.
When doing hypothesis testing, what is your important assumption?
That one of the hypothesizes (H0 or HA) is true.
Type 1 error?
If null hypothesis is true and the decision at the end of the hypothesis test is to reject the null hypothesis.
Type II error?
If null hypothesis is false and the decision at the end of the hypothesis test is to fail to reject the null hypothesis.
Significance level of a test (alpha)?
Predetermined probability that the test will commit a type I error
Power function (1- beta(theta))?
Predetermined probability that that test will reject H0.
If we don’t know whether H0 or HA is correct, what do we want to do when designing our test?
Keep the chance as small as possible that type I and type II error will happen.
Test statistic?
Statistic to define the rejection and acceptance regions.
Critical value?
A number on the boundary between the rejection region and the acceptance region.
Test statistic is measuring?
Difference between observed pe and pe if H0 is true
Left-tailed?
Theta < theta_0 -> rejection region is less and far from null value.
Right-tailed?
Theta > theta_0 -> point estimate is far and bigger from the null value.
