Kolmogorov-Smirnov test in SciPy - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is the Kolmogorov-Smirnov test used for?

It is a statistical test used to compare a sample distribution with a reference probability distribution, or to compare two sample distributions, to check if they differ significantly.

beginner

What does the KS test statistic represent?

The KS test statistic is the maximum distance between the cumulative distribution functions (CDFs) of the two distributions being compared.

beginner

In scipy, which function is used to perform the Kolmogorov-Smirnov test for one sample against a distribution?

The function is scipy.stats.kstest(). It compares a sample against a given distribution.

beginner

What does a small p-value in the KS test indicate?

A small p-value means there is strong evidence that the sample distribution is different from the reference distribution.

beginner

How do you interpret the KS test result when comparing two samples?

If the p-value is large, it suggests the two samples come from the same distribution; if small, they likely come from different distributions.

What does the Kolmogorov-Smirnov test compare?

A Two distributions or a sample against a distribution

B Means of two samples

C Variances of two samples

D Correlation between two variables

The KS test compares the cumulative distributions of two samples or a sample and a reference distribution.

Which scipy function performs the KS test for one sample?

A scipy.stats.ttest_1samp

B scipy.stats.chisquare

C scipy.stats.kstest

D scipy.stats.ks_2samp

scipy.stats.kstest compares one sample against a reference distribution.

What does a KS test statistic measure?

A Difference in means

B Maximum difference between CDFs

C Sum of squared errors

D Correlation coefficient

The KS statistic is the largest vertical distance between two cumulative distribution functions.

If the KS test p-value is 0.01, what does it mean?

A Distributions are likely different

B Distributions are likely the same

C Test is inconclusive

D Sample size is too small

A small p-value (like 0.01) suggests strong evidence against the null hypothesis that distributions are the same.

Which test compares two samples directly using KS test?

A scipy.stats.mannwhitneyu

B scipy.stats.kstest

C scipy.stats.ttest_ind

D scipy.stats.ks_2samp

scipy.stats.ks_2samp compares two independent samples using the KS test.

Explain how the Kolmogorov-Smirnov test works and what it measures.

Think about how you compare two shapes of data distributions.

Describe how to use scipy to perform a KS test on a sample against a normal distribution.

Remember the function name and what arguments it needs.