h = ttest2(x, y)
h = ttest2(x, y, 'Alpha', alpha)
h = ttest2(x, y, 'Tail', tail)
h = ttest2(x, y, 'Vartype', vartype)
h = ttest2(x, y, 'Dim', dim)
[h, p, ci, stats] = ttest2(...)
| Parameter | Description |
|---|---|
| x | real numeric array: first sample. |
| y | real numeric array: second sample. |
| alpha | scalar in (0,1), 0.05 by default: significance level. |
| tail | 'both', 'right', or 'left'. |
| vartype | 'equal' by default or 'unequal' for Welch's test. |
| dim | positive integer: dimension to operate along. |
| Parameter | Description |
|---|---|
| h | logical array: test decision. |
| p | array: p-values. |
| ci | 2-row array: confidence intervals for the mean difference. |
| stats | structure with tstat, df, and sd fields. |
ttest2 performs a two-sample t-test along the first non-singleton dimension unless Dim is specified.
NaN values are omitted independently from each tested sample.
x = [10 11 13 15 18];
y = [7 8 8 9];
[h, p, ci, stats] = ttest2(x, y, 'Vartype', 'unequal', 'Tail', 'right');| Version | Description |
|---|---|
| 2.0.0 | initial version |