Skip to content

Commit 4152dbe

Browse files
DivyanshuVortexDivyanshuVortex
authored
feat: add stats/base/ndarray/variance
PR-URL: #9828 Reviewed-by: Athan Reines <kgryte@gmail.com>
1 parent 3d6cd7d commit 4152dbe

11 files changed

Lines changed: 969 additions & 0 deletions

File tree

‎lib/node_modules/@stdlib/stats/base/ndarray/variance/README.md‎

Lines changed: 186 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,186 @@
1+
<!--
2+
3+
@license Apache-2.0
4+
5+
Copyright (c) 2026 The Stdlib Authors.
6+
7+
Licensed under the Apache License, Version 2.0 (the "License");
8+
you may not use this file except in compliance with the License.
9+
You may obtain a copy of the License at
10+
11+
http://www.apache.org/licenses/LICENSE-2.0
12+
13+
Unless required by applicable law or agreed to in writing, software
14+
distributed under the License is distributed on an "AS IS" BASIS,
15+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
16+
See the License for the specific language governing permissions and
17+
limitations under the License.
18+
19+
-->
20+
21+
# variance
22+
23+
> Calculate the [variance][variance] of a one-dimensional ndarray.
24+
25+
<section class="intro">
26+
27+
The population [variance][variance] of a finite size population of size `N` is given by
28+
29+
<!-- <equation class="equation" label="eq:population_variance" align="center" raw="\sigma^2 = \frac{1}{N} \sum_{i=0}^{N-1} (x_i - \mu)^2" alt="Equation for the population variance."> -->
30+
31+
```math
32+
\sigma^2 = \frac{1}{N} \sum_{i=0}^{N-1} (x_i - \mu)^2
33+
```
34+
35+
<!-- <div class="equation" align="center" data-raw-text="\sigma^2 = \frac{1}{N} \sum_{i=0}^{N-1} (x_i - \mu)^2}" data-equation="eq:population_variance">
36+
<img src="https://cdn.jsdelivr.net/gh/stdlib-js/stdlib@08ca32895957967bd760a4fe02d61762432a0b72/lib/node_modules/@stdlib/stats/strided/variance/docs/img/equation_population_variance.svg" alt="Equation for the population variance.">
37+
<br>
38+
</div> -->
39+
40+
<!-- </equation> -->
41+
42+
where the population mean is given by
43+
44+
<!-- <equation class="equation" label="eq:population_mean" align="center" raw="\mu = \frac{1}{N} \sum_{i=0}^{N-1} x_i" alt="Equation for the population mean."> -->
45+
46+
```math
47+
\mu = \frac{1}{N} \sum_{i=0}^{N-1} x_i
48+
```
49+
50+
<!-- <div class="equation" align="center" data-raw-text="\mu = \frac{1}{N} \sum_{i=0}^{N-1} x_i" data-equation="eq:population_mean">
51+
<img src="https://cdn.jsdelivr.net/gh/stdlib-js/stdlib@08ca32895957967bd760a4fe02d61762432a0b72/lib/node_modules/@stdlib/stats/strided/variance/docs/img/equation_population_mean.svg" alt="Equation for the population mean.">
52+
<br>
53+
</div> -->
54+
55+
<!-- </equation> -->
56+
57+
Often in the analysis of data, the true population [variance][variance] is not known _a priori_ and must be estimated from a sample drawn from the population distribution. If one attempts to use the formula for the population [variance][variance], the result is biased and yields an **uncorrected sample variance**. To compute a **corrected sample variance** for a sample of size `n`,
58+
59+
<!-- <equation class="equation" label="eq:corrected_sample_variance" align="center" raw="s^2 = \frac{1}{n-1} \sum_{i=0}^{n-1} (x_i - \bar{x})^2" alt="Equation for computing a corrected sample variance."> -->
60+
61+
```math
62+
s^2 = \frac{1}{n-1} \sum_{i=0}^{n-1} (x_i - \bar{x})^2
63+
```
64+
65+
<!-- <div class="equation" align="center" data-raw-text="s^2 = \frac{1}{n-1} \sum_{i=0}^{n-1} (x_i - \bar{x})^2}" data-equation="eq:corrected_sample_variance">
66+
<img src="https://cdn.jsdelivr.net/gh/stdlib-js/stdlib@08ca32895957967bd760a4fe02d61762432a0b72/lib/node_modules/@stdlib/stats/strided/variance/docs/img/equation_corrected_sample_variance.svg" alt="Equation for computing a corrected sample variance.">
67+
<br>
68+
</div> -->
69+
70+
<!-- </equation> -->
71+
72+
where the sample mean is given by
73+
74+
<!-- <equation class="equation" label="eq:sample_mean" align="center" raw="\bar{x} = \frac{1}{n} \sum_{i=0}^{n-1} x_i" alt="Equation for the sample mean."> -->
75+
76+
```math
77+
\bar{x} = \frac{1}{n} \sum_{i=0}^{n-1} x_i
78+
```
79+
80+
<!-- <div class="equation" align="center" data-raw-text="\bar{x} = \frac{1}{n} \sum_{i=0}^{n-1} x_i" data-equation="eq:sample_mean">
81+
<img src="https://cdn.jsdelivr.net/gh/stdlib-js/stdlib@08ca32895957967bd760a4fe02d61762432a0b72/lib/node_modules/@stdlib/stats/strided/variance/docs/img/equation_sample_mean.svg" alt="Equation for the sample mean.">
82+
<br>
83+
</div> -->
84+
85+
<!-- </equation> -->
86+
87+
The use of the term `n-1` is commonly referred to as Bessel's correction. Note, however, that applying Bessel's correction can increase the mean squared error between the sample variance and population variance. Depending on the characteristics of the population distribution, other correction factors (e.g., `n-1.5`, `n+1`, etc) can yield better estimators.
88+
89+
</section>
90+
91+
<!-- /.intro -->
92+
93+
<section class="usage">
94+
95+
## Usage
96+
97+
```javascript
98+
var variance = require( '@stdlib/stats/base/ndarray/variance' );
99+
```
100+
101+
#### variance( arrays )
102+
103+
Computes the [variance][variance] of a one-dimensional ndarray.
104+
105+
```javascript
106+
var ndarray = require( '@stdlib/ndarray/base/ctor' );
107+
var scalar2ndarray = require( '@stdlib/ndarray/from-scalar' );
108+
109+
var opts = {
110+
'dtype': 'generic'
111+
};
112+
113+
var xbuf = [ 1.0, -2.0, 2.0 ];
114+
var x = new ndarray( opts.dtype, xbuf, [ 3 ], [ 1 ], 0, 'row-major' );
115+
var correction = scalar2ndarray( 1.0, opts );
116+
117+
var v = variance( [ x, correction ] );
118+
// returns ~4.3333
119+
```
120+
121+
The function has the following parameters:
122+
123+
- **arrays**: array-like object containing two elements: a one-dimensional input ndarray and a zero-dimensional ndarray specifying the degrees of freedom adjustment. Providing a non-zero degrees of freedom adjustment has the effect of adjusting the divisor during the calculation of the [variance][variance] according to `N-c` where `N` is the number of elements in the input ndarray and `c` corresponds to the provided degrees of freedom adjustment. When computing the [variance][variance] of a population, setting this parameter to `0` is the standard choice (i.e., the provided array contains data constituting an entire population). When computing the corrected sample [variance][variance], setting this parameter to `1` is the standard choice (i.e., the provided array contains data sampled from a larger population; this is commonly referred to as Bessel's correction).
124+
125+
</section>
126+
127+
<!-- /.usage -->
128+
129+
<section class="notes">
130+
131+
## Notes
132+
133+
- If provided an empty one-dimensional ndarray, the function returns `NaN`.
134+
- If `N - c` is less than or equal to `0` (where `N` corresponds to the number of elements in the input ndarray and `c` corresponds to the provided degrees of freedom adjustment), the function returns `NaN`.
135+
136+
</section>
137+
138+
<!-- /.notes -->
139+
140+
<section class="examples">
141+
142+
## Examples
143+
144+
<!-- eslint no-undef: "error" -->
145+
146+
```javascript
147+
var discreteUniform = require( '@stdlib/random/array/discrete-uniform' );
148+
var ndarray = require( '@stdlib/ndarray/base/ctor' );
149+
var scalar2ndarray = require( '@stdlib/ndarray/from-scalar' );
150+
var ndarray2array = require( '@stdlib/ndarray/to-array' );
151+
var variance = require( '@stdlib/stats/base/ndarray/variance' );
152+
153+
var opts = {
154+
'dtype': 'float64'
155+
};
156+
157+
var xbuf = discreteUniform( 10, -50, 50, opts );
158+
var x = new ndarray( opts.dtype, xbuf, [ xbuf.length ], [ 1 ], 0, 'row-major' );
159+
console.log( ndarray2array( x ) );
160+
161+
var correction = scalar2ndarray( 1.0, opts );
162+
var v = variance( [ x, correction ] );
163+
console.log( v );
164+
```
165+
166+
</section>
167+
168+
<!-- /.examples -->
169+
170+
<!-- Section for related `stdlib` packages. Do not manually edit this section, as it is automatically populated. -->
171+
172+
<section class="related">
173+
174+
</section>
175+
176+
<!-- /.related -->
177+
178+
<!-- Section for all links. Make sure to keep an empty line after the `section` element and another before the `/section` close. -->
179+
180+
<section class="links">
181+
182+
[variance]: https://en.wikipedia.org/wiki/Variance
183+
184+
</section>
185+
186+
<!-- /.links -->

‎lib/node_modules/@stdlib/stats/base/ndarray/variance/benchmark/benchmark.js‎

Lines changed: 112 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,112 @@
1+
/**
2+
* @license Apache-2.0
3+
*
4+
* Copyright (c) 2026 The Stdlib Authors.
5+
*
6+
* Licensed under the Apache License, Version 2.0 (the "License");
7+
* you may not use this file except in compliance with the License.
8+
* You may obtain a copy of the License at
9+
*
10+
* http://www.apache.org/licenses/LICENSE-2.0
11+
*
12+
* Unless required by applicable law or agreed to in writing, software
13+
* distributed under the License is distributed on an "AS IS" BASIS,
14+
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
15+
* See the License for the specific language governing permissions and
16+
* limitations under the License.
17+
*/
18+
19+
'use strict';
20+
21+
// MODULES //
22+
23+
var bench = require( '@stdlib/bench' );
24+
var uniform = require( '@stdlib/random/array/uniform' );
25+
var isnan = require( '@stdlib/math/base/assert/is-nan' );
26+
var pow = require( '@stdlib/math/base/special/pow' );
27+
var ndarray = require( '@stdlib/ndarray/base/ctor' );
28+
var scalar2ndarray = require( '@stdlib/ndarray/from-scalar' );
29+
var format = require( '@stdlib/string/format' );
30+
var pkg = require( './../package.json' ).name;
31+
var variance = require( './../lib' );
32+
33+
34+
// VARIABLES //
35+
36+
var options = {
37+
'dtype': 'generic'
38+
};
39+
40+
41+
// FUNCTIONS //
42+
43+
/**
44+
* Creates a benchmark function.
45+
*
46+
* @private
47+
* @param {PositiveInteger} len - array length
48+
* @returns {Function} benchmark function
49+
*/
50+
function createBenchmark( len ) {
51+
var correction;
52+
var xbuf;
53+
var x;
54+
55+
xbuf = uniform( len, -10.0, 10.0, options );
56+
x = new ndarray( options.dtype, xbuf, [ len ], [ 1 ], 0, 'row-major' );
57+
correction = scalar2ndarray( 1.0, options );
58+
59+
return benchmark;
60+
61+
/**
62+
* Benchmark function.
63+
*
64+
* @private
65+
* @param {Benchmark} b - benchmark instance
66+
*/
67+
function benchmark( b ) {
68+
var v;
69+
var i;
70+
71+
b.tic();
72+
for ( i = 0; i < b.iterations; i++ ) {
73+
v = variance( [ x, correction ] );
74+
if ( isnan( v ) ) {
75+
b.fail( 'should not return NaN' );
76+
}
77+
}
78+
b.toc();
79+
if ( isnan( v ) ) {
80+
b.fail( 'should not return NaN' );
81+
}
82+
b.pass( 'benchmark finished' );
83+
b.end();
84+
}
85+
}
86+
87+
88+
// MAIN //
89+
90+
/**
91+
* Main execution sequence.
92+
*
93+
* @private
94+
*/
95+
function main() {
96+
var len;
97+
var min;
98+
var max;
99+
var f;
100+
var i;
101+
102+
min = 1; // 10^min
103+
max = 6; // 10^max
104+
105+
for ( i = min; i <= max; i++ ) {
106+
len = pow( 10, i );
107+
f = createBenchmark( len );
108+
bench( format( '%s:len=%d', pkg, len ), f );
109+
}
110+
}
111+
112+
main();

‎lib/node_modules/@stdlib/stats/base/ndarray/variance/docs/img/equation_sample_mean.svg‎

Lines changed: 43 additions & 0 deletions
Loading

‎lib/node_modules/@stdlib/stats/base/ndarray/variance/docs/repl.txt‎

Lines changed: 53 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,53 @@
1+
2+
{{alias}}( arrays )
3+
Computes the variance of a one-dimensional ndarray.
4+
5+
If provided an empty one-dimensional ndarray, the function returns `NaN`.
6+
7+
If `N - c` is less than or equal to `0` (where `N` corresponds to the number
8+
of elements in the input ndarray and `c` corresponds to the provided degrees
9+
of freedom adjustment), the function returns `NaN`.
10+
11+
Parameters
12+
----------
13+
arrays: ArrayLikeObject<ndarray>
14+
Array-like object containing two elements: a one-dimensional input
15+
ndarray and a zero-dimensional ndarray specifying the degrees of freedom
16+
adjustment. Providing a non-zero degrees of freedom adjustment has the
17+
effect of adjusting the divisor during the calculation of the variance
18+
according to `N-c` where `N` is the number of elements in the input
19+
ndarray and `c` corresponds to the provided degrees of freedom
20+
adjustment. When computing the variance of a population, setting this
21+
parameter to `0` is the standard choice (i.e., the provided array
22+
contains data constituting an entire population). When computing the
23+
corrected sample variance, setting this parameter to `1` is the standard
24+
choice (i.e., the provided array contains data sampled from a larger
25+
population; this is commonly referred to as Bessel's correction).
26+
27+
Returns
28+
-------
29+
out: number
30+
The variance.
31+
32+
Examples
33+
--------
34+
// Create input ndarray:
35+
> var xbuf = [ 1.0, -2.0, 2.0 ];
36+
> var dt = 'generic';
37+
> var sh = [ xbuf.length ];
38+
> var st = [ 1 ];
39+
> var oo = 0;
40+
> var ord = 'row-major';
41+
> var x = new {{alias:@stdlib/ndarray/ctor}}( dt, xbuf, sh, st, oo, ord );
42+
43+
// Create correction ndarray:
44+
> var opts = { 'dtype': dt };
45+
> var correction = {{alias:@stdlib/ndarray/from-scalar}}( 1.0, opts );
46+
47+
// Compute the variance:
48+
> {{alias}}( [ x, correction ] )
49+
~4.3333
50+
51+
See Also
52+
--------
53+

0 commit comments

Comments
 (0)