One-Sample Kolmogorov-Smirnov Test: Simulation
When certain parameters of the distribution have to be estimated from the sample, the Kolmogorov-Smirnov test no longer applies. In these instances, the Lilliefors test statistic can be used to estimate the p-value by using the Monte Carlo sampling for testing normality with mean and variance unknown. The Lilliefors test applies to the three continuous distributions (Normal, Exponential, and Uniform). Note that the test does not apply if the underlying distribution is discrete (Poisson). The test is only defined for one-sample inference when the corresponding distribution parameters are not specified.
- Monte Carlo Simulation Parameters
-
- Confidence level
- This optional setting resets the confidence interval level that is estimated by the Kolmogorov-Smirnov test when using the Monte Carlo simulations. The value must be between 0 and 100. The default setting is 99.
- Number of samples
- This optional setting resets the number of replicates that the Lilliefors test uses for the Monte Carlo sampling. The value must be a single integer between 10000 and the largest number of samples value. The default value is 10000.
- Suppress the Monte Carlo results for the normal distribution
- This optional setting suppresses the Monte Carlo sampling for the normal distribution results. By default, the setting is not selected (which means both the existing asymptotic results and the Lilliefors test results, that are based on the Monte Carlo sampling, are presented).
Specifying simulation settings for the One-Sample Kolmogorov-Smirnov Test
This feature requires the Statistics Base option.
- From the menus choose:
- In the One-Sample Kolmogorov-Smirnov Test dialog box, select the desired test variables, select at least one Test Distribution method, and then click Simulation....