|
| 1 | +<!-- |
| 2 | +
|
| 3 | +@license Apache-2.0 |
| 4 | +
|
| 5 | +Copyright (c) 2025 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 | +# incrnanskewness |
| 22 | + |
| 23 | +> Compute a [corrected sample skewness][sample-skewness] incrementally, ignoring `NaN` values. |
| 24 | +
|
| 25 | +<section class="intro"> |
| 26 | + |
| 27 | +The [skewness][sample-skewness] for a random variable `X` is defined as |
| 28 | + |
| 29 | +<!-- <equation class="equation" label="eq:skewness" align="center" raw="\operatorname{Skewness}[X] = \mathrm{E}\biggl[ \biggl( \frac{X - \mu}{\sigma} \biggr)^3 \biggr]" alt="Equation for skewness."> --> |
| 30 | + |
| 31 | +```math |
| 32 | +\mathop{\mathrm{Skewness}}[X] = \mathrm{E}\biggl[ \biggl( \frac{X - \mu}{\sigma} \biggr)^3 \biggr] |
| 33 | +``` |
| 34 | + |
| 35 | +<!-- <div class="equation" align="center" data-raw-text="\operatorname{Skewness}[X] = \mathrm{E}\biggl[ \biggl( \frac{X - \mu}{\sigma} \biggr)^3 \biggr]" data-equation="eq:skewness"> |
| 36 | + <img src="https://cdn.jsdelivr.net/gh/stdlib-js/stdlib@49d8cabda84033d55d7b8069f19ee3dd8b8d1496/lib/node_modules/@stdlib/stats/incr/nanskewness/docs/img/equation_skewness.svg" alt="Equation for skewness."> |
| 37 | + <br> |
| 38 | +</div> --> |
| 39 | + |
| 40 | +<!-- </equation> --> |
| 41 | + |
| 42 | +For a sample of `n` values, the [sample skewness][sample-skewness] is |
| 43 | + |
| 44 | +<!-- <equation class="equation" label="eq:sample_skewness" align="center" raw="b_1 = \frac{m_3}{s^3} = \frac{\frac{1}{n} \sum_{i=0}^{n-1} (x_i - \bar{x})^3}{\biggl( \frac{1}{n-1} \sum_{i=0}^{n-1} (x_i - \bar{x})^2 \biggr)^{3/2}}" alt="Equation for the sample skewness."> --> |
| 45 | + |
| 46 | +```math |
| 47 | +b_1 = \frac{m_3}{s^3} = \frac{\frac{1}{n} \sum_{i=0}^{n-1} (x_i - \bar{x})^3}{\biggl( \frac{1}{n-1} \sum_{i=0}^{n-1} (x_i - \bar{x})^2 \biggr)^{3/2}} |
| 48 | +``` |
| 49 | + |
| 50 | +<!-- <div class="equation" align="center" data-raw-text="b_1 = \frac{m_3}{s^3} = \frac{\frac{1}{n} \sum_{i=0}^{n-1} (x_i - \bar{x})^3}{\biggl( \frac{1}{n-1} \sum_{i=0}^{n-1} (x_i - \bar{x})^2 \biggr)^{3/2}}" data-equation="eq:sample_skewness"> |
| 51 | + <img src="https://cdn.jsdelivr.net/gh/stdlib-js/stdlib@49d8cabda84033d55d7b8069f19ee3dd8b8d1496/lib/node_modules/@stdlib/stats/incr/nanskewness/docs/img/equation_sample_skewness.svg" alt="Equation for the sample skewness."> |
| 52 | + <br> |
| 53 | +</div> --> |
| 54 | + |
| 55 | +<!-- </equation> --> |
| 56 | + |
| 57 | +where `m_3` is the sample third central moment and `s` is the sample standard deviation. |
| 58 | + |
| 59 | +An alternative definition for the [sample skewness][sample-skewness] which includes an adjustment factor (and is the implemented definition) is |
| 60 | + |
| 61 | +<!-- <equation class="equation" label="eq:adjusted_sample_skewness" align="center" raw="G_1 = \frac{n^2}{(n-1)(n-2)} \frac{m_3}{s^3} = \frac{\sqrt{n(n-1)}}{n-2} \frac{\frac{1}{n} \sum_{i=0}^{n-1} (x_i - \bar{x})^3}{\biggl( \frac{1}{n} \sum_{i=0}^{n-1} (x_i - \bar{x})^2 \biggr)^{3/2}}" alt="Equation for the adjusted sample skewness."> --> |
| 62 | + |
| 63 | +```math |
| 64 | +G_1 = \frac{n^2}{(n-1)(n-2)} \frac{m_3}{s^3} = \frac{\sqrt{n(n-1)}}{n-2} \frac{\frac{1}{n} \sum_{i=0}^{n-1} (x_i - \bar{x})^3}{\biggl( \frac{1}{n} \sum_{i=0}^{n-1} (x_i - \bar{x})^2 \biggr)^{3/2}} |
| 65 | +``` |
| 66 | + |
| 67 | +<!-- <div class="equation" align="center" data-raw-text="G_1 = \frac{n^2}{(n-1)(n-2)} \frac{m_3}{s^3} = \frac{\sqrt{n(n-1)}}{n-2} \frac{\frac{1}{n} \sum_{i=0}^{n-1} (x_i - \bar{x})^3}{\biggl( \frac{1}{n} \sum_{i=0}^{n-1} (x_i - \bar{x})^2 \biggr)^{3/2}}" data-equation="eq:adjusted_sample_skewness"> |
| 68 | + <img src="https://cdn.jsdelivr.net/gh/stdlib-js/stdlib@49d8cabda84033d55d7b8069f19ee3dd8b8d1496/lib/node_modules/@stdlib/stats/incr/nanskewness/docs/img/equation_adjusted_sample_skewness.svg" alt="Equation for the adjusted sample skewness."> |
| 69 | + <br> |
| 70 | +</div> --> |
| 71 | + |
| 72 | +<!-- </equation> --> |
| 73 | + |
| 74 | +</section> |
| 75 | + |
| 76 | +<!-- /.intro --> |
| 77 | + |
| 78 | +<section class="usage"> |
| 79 | + |
| 80 | +## Usage |
| 81 | + |
| 82 | +```javascript |
| 83 | +var incrnanskewness = require( '@stdlib/stats/incr/nanskewness' ); |
| 84 | +``` |
| 85 | + |
| 86 | +#### incrnanskewness() |
| 87 | + |
| 88 | +Returns an accumulator function which incrementally computes a [corrected sample skewness][sample-skewness], ignoring `NaN` values. |
| 89 | + |
| 90 | +```javascript |
| 91 | +var accumulator = incrnanskewness(); |
| 92 | +``` |
| 93 | + |
| 94 | +#### accumulator( \[x] ) |
| 95 | + |
| 96 | +If provided an input value `x`, the accumulator function returns an updated [corrected sample skewness][sample-skewness]. If not provided an input value `x`, the accumulator function returns the current [corrected sample skewness][sample-skewness]. |
| 97 | + |
| 98 | +```javascript |
| 99 | +var accumulator = incrnanskewness(); |
| 100 | + |
| 101 | +var skewness = accumulator(); |
| 102 | +// returns null |
| 103 | + |
| 104 | +skewness = accumulator( 2.0 ); |
| 105 | +// returns null |
| 106 | + |
| 107 | +skewness = accumulator( -5.0 ); |
| 108 | +// returns null |
| 109 | + |
| 110 | +skewness = accumulator( -10.0 ); |
| 111 | +// returns ~0.492 |
| 112 | + |
| 113 | +skewness = accumulator( NaN ); |
| 114 | +// returns ~0.492 |
| 115 | + |
| 116 | +skewness = accumulator(); |
| 117 | +// returns ~0.492 |
| 118 | +``` |
| 119 | + |
| 120 | +</section> |
| 121 | + |
| 122 | +<!-- /.usage --> |
| 123 | + |
| 124 | +<section class="notes"> |
| 125 | + |
| 126 | +## Notes |
| 127 | + |
| 128 | +- Input values are **not** type checked. If non-numeric inputs are possible, you are advised to type check and handle accordingly **before** passing the value to the accumulator function. |
| 129 | + |
| 130 | +</section> |
| 131 | + |
| 132 | +<!-- /.notes --> |
| 133 | + |
| 134 | +<section class="examples"> |
| 135 | + |
| 136 | +## Examples |
| 137 | + |
| 138 | +<!-- eslint no-undef: "error" --> |
| 139 | + |
| 140 | +```javascript |
| 141 | +var uniform = require( '@stdlib/random/base/uniform' ); |
| 142 | +var bernoulli = require( '@stdlib/random/base/bernoulli' ); |
| 143 | +var incrnanskewness = require( '@stdlib/stats/incr/nanskewness' ); |
| 144 | + |
| 145 | +// Initialize an accumulator: |
| 146 | +var accumulator = incrnanskewness(); |
| 147 | + |
| 148 | +// For each simulated datum, update the corrected sample skewness... |
| 149 | +var i; |
| 150 | +for ( i = 0; i < 100; i++ ) { |
| 151 | + accumulator( ( bernoulli( 0.8 ) < 1 ) ? NaN : uniform( 0.0, 100.0 ) ); |
| 152 | +} |
| 153 | +console.log( accumulator() ); |
| 154 | +``` |
| 155 | + |
| 156 | +</section> |
| 157 | + |
| 158 | +<!-- /.examples --> |
| 159 | + |
| 160 | +<section class="references"> |
| 161 | + |
| 162 | +</section> |
| 163 | + |
| 164 | +<!-- /.references --> |
| 165 | + |
| 166 | +<!-- Section for related `stdlib` packages. Do not manually edit this section, as it is automatically populated. --> |
| 167 | + |
| 168 | +<section class="related"> |
| 169 | + |
| 170 | +</section> |
| 171 | + |
| 172 | +<!-- /.related --> |
| 173 | + |
| 174 | +<!-- Section for all links. Make sure to keep an empty line after the `section` element and another before the `/section` close. --> |
| 175 | + |
| 176 | +<section class="links"> |
| 177 | + |
| 178 | +[sample-skewness]: https://en.wikipedia.org/wiki/Skewness |
| 179 | + |
| 180 | +</section> |
| 181 | + |
| 182 | +<!-- /.links --> |
0 commit comments