mirror of
https://github.com/ClickHouse/ClickHouse.git
synced 2024-11-13 02:53:38 +00:00
363aefb42e
* studentTTest function description * welchttest description, links in studentttest * mannWhitneyUTest description * Text changed slightly * Text changed slightly * Text changed slightly * Text changed slightly * - First header is first level. - sample_index type set to UInt64. - Return values renamed. - mannWhitneyUTest: - added continuity_correction parameter - optional parameters wrapped in braces [ ] - description refactored - example SQL placed on single line * Improved formulations (hopefully) General data type names used (Integer, Float, Decimal) Removed names from return values * Fixes * Optional parameters stated explicitly * New line at the end of files * Russian function description * Files moved in the right folder * Translated 'Original article' to russian * Typos are fixed * fixed alowed parameter value: "two sided" -> "two-sided" * duplicate word "require" is removed from exception messages * mannWhitneyUTest description changed Added Tuple link in return values "samples from two populations" added in short descriptions * Russian translation updated * fix Co-authored-by: Alexey Smirnov <alexey-sm@yandex-team.ru>
2.7 KiB
2.7 KiB
| toc_priority | toc_title |
|---|---|
| 310 | mannWhitneyUTest |
mannWhitneyUTest
Applies the Mann-Whitney rank test to samples from two populations.
Syntax
mannWhitneyUTest[(alternative[, continuity_correction])](sample_data, sample_index)
Values of both samples are in the sample_data column. If sample_index equals to 0 then the value in that row belongs to the sample from the first population. Otherwise it belongs to the sample from the second population.
The null hypothesis is that two populations are stochastically equal. Also one-sided hypothesises can be tested. This test does not assume that data have normal distribution.
Parameters
alternative— alternative hypothesis. (Optional, default:'two-sided'.) String.'two-sided';'greater';'less'.
continuity_correction- if not 0 then continuity correction in the normal approximation for the p-value is applied. (Optional, default: 1.) UInt64.sample_data— sample data. Integer, Float or Decimal.sample_index— sample index. Integer.
Returned values
Tuple with two elements:
Example
Input table:
┌─sample_data─┬─sample_index─┐
│ 10 │ 0 │
│ 11 │ 0 │
│ 12 │ 0 │
│ 1 │ 1 │
│ 2 │ 1 │
│ 3 │ 1 │
└─────────────┴──────────────┘
Query:
SELECT mannWhitneyUTest('greater')(sample_data, sample_index) FROM mww_ttest;
Result:
┌─mannWhitneyUTest('greater')(sample_data, sample_index)─┐
│ (9,0.04042779918503192) │
└────────────────────────────────────────────────────────┘
See Also