Statistics and Machine Learning Toolbox™ provides parametric and nonparametric hypothesis tests to help you determine if your sample data comes from a population with particular characteristics.
Distribution tests , such as Anderson-Darling and one-sample Kolmogorov-Smirnov, test whether sample data comes from a population with a particular distribution. Test whether two sets of sample data have the same distribution using tests such as two-sample Kolmogorov-Smirnov.
Location tests, such as z-test and one-sample t-test, test whether sample data comes from a population with a particular mean or median. Test two or more sets of sample data for the same location value using a two-sample t-test or multiple comparison test.
Dispersion tests, such as Chi-square variance, test whether sample data comes from a population with a particular variance. Compare the variances of two or more sample data sets using a two-sample F-test or multiple-sample test.
Determine additional features of sample data by cross-tabulating, conducting a run test for randomness, and determine the sample size and power for a hypothesis test.
|Chi-square goodness-of-fit test|
|Fisher’s exact test|
|One-sample Kolmogorov-Smirnov test|
|Two-sample Kolmogorov-Smirnov test|
|Run test for randomness|
Hypothesis testing is a common method of drawing inferences about a population based on statistical evidence from a sample.
All hypothesis tests share the same basic terminology and structure.
Different hypothesis tests make different assumptions about the distribution of the random variable being sampled in the data.
This example shows how to determine the number of samples or observations needed to carry out a statistical test.
Tests of distributions and statistics