Documentation

This is machine translation

Translated by Microsoft
Mouse over text to see original. Click the button below to return to the English verison of the page.

Principles of Hypothesis Testing

Hypothesis (goodness-of-fit) testing is a common method that uses statistical evidence from a sample to draw a conclusion about a population. In hypothesis testing, you assert a particular statement (a null hypothesis) and try to find evidence to support or reject that statement. A null hypothesis is an assumption about a population that you would like to test. It is "null" in the sense that it often represents a status-quo belief, such as the absence of a characteristic or the lack of an effect. You can formalize it by asserting that a population parameter, or a combination of population parameters, has a certain value. MuPAD® enables you to test the following null hypotheses:

  • The data has the distribution function f. If f is a cumulative distribution function (CDF), you can use the classical chi-square goodness-of-fit test or the Kolmogorov-Smirnov test. If f is probability density function (PDF), a discrete probability function (PF), or an arbitrary distribution function, use the classical chi-square goodness-of-fit test.

  • The data has a normal distribution function with a particular mean and a particular variance. For cumulative distribution functions, use the classical chi-square goodness-of-fit test or the Kolmogorov-Smirnov test. For other distribution functions, use the classical chi-square goodness-of-fit test.

  • The data has a normal distribution function (with unknown mean and variance). Use the Shapiro-Wilk goodness-of-fit test for this hypothesis.

  • The mean of the data is larger than some particular value. Use the t-Test for this hypothesis.

The main result returned by the hypothesis tests is the p-value (PValue). The p-value of a test indicates the probability, under the null hypothesis, of obtaining a value of the test statistic as extreme or more extreme than the value computed from the sample. If the p-value is larger than the significance level (stated and agreed upon before the test), the null hypothesis passes the test. A typical value of a significance level is 0.05. P-values below a significance level provide strong evidence for rejecting the null hypothesis.

Was this topic helpful?