MathematicsCode 9709
Hypothesis tests
What this topic asks you to be able to do
- understand what a hypothesis test is and how one-tailed tests differ from two-tailed ones, and know what is meant by the null and alternative hypotheses, significance level, rejection (or critical) region, acceptance region and test statistic
- set up hypotheses and carry out a hypothesis test based on one observation from a population with a binomial or Poisson distribution, working out probabilities directly or, where appropriate, using a normal approximation for the binomial or Poisson distribution
- set up hypotheses and perform a test on the population mean for situations in which the population follows a normal distribution with known variance, or in which a large sample is taken
- understand what is meant by Type I error and Type II error when hypothesis tests are considered
- work out the probability of a Type I or a Type II error in given situations, for tests that use a normal distribution or that evaluate binomial or Poisson probabilities directly