h = kstest(x)
h = kstest(x, 'CDF', cdf)
h = kstest(x, 'CDF', cdfFunction)
[h, p, ksstat, cv] = kstest(..., 'Alpha', alpha, 'Tail', tail)
| Parameter | Description |
|---|---|
| x | real vector: sample data. |
| cdf | two-column matrix defining x values and cumulative probabilities, function handle evaluated at the sorted sample values, or object with a cdf method. |
| cdfFunction | function handle returning cumulative probabilities for each input sample value. |
| alpha | scalar in (0,1), 0.05 by default: significance level. |
| tail | 'unequal', 'larger', or 'smaller'. |
| Parameter | Description |
|---|---|
| h | logical scalar: test decision. |
| p | p-value. |
| ksstat | test statistic. |
| cv | critical value. |
kstest compares the empirical distribution of x with the standard normal distribution or a user-supplied cumulative distribution.
NaN sample values are omitted before sorting and computing the empirical distribution.
x = [-1.2 -0.4 0.1 0.3 0.8];
[h, p, ksstat, cv] = kstest(x);
cdf = [-2 0; -1 0.2; 0 0.5; 1 0.8; 2 1];
h2 = kstest(x, 'CDF', cdf);
h3 = kstest([0 1 2], 'CDF', @(z) 0.2 + 0.3 .* z);| Version | Description |
|---|---|
| 2.0.0 | initial version |