diff --git a/lib/node_modules/@stdlib/stats/base/nanvariancepn/README.md b/lib/node_modules/@stdlib/stats/base/nanvariancepn/README.md index d66dfc5a8017..dcb9fb7388b6 100644 --- a/lib/node_modules/@stdlib/stats/base/nanvariancepn/README.md +++ b/lib/node_modules/@stdlib/stats/base/nanvariancepn/README.md @@ -98,9 +98,9 @@ The use of the term `n-1` is commonly referred to as Bessel's correction. Note, var nanvariancepn = require( '@stdlib/stats/base/nanvariancepn' ); ``` -#### nanvariancepn( N, correction, x, stride ) +#### nanvariancepn( N, correction, x, strideX ) -Computes the [variance][variance] of a strided array `x` ignoring `NaN` values and using a two-pass algorithm. +Computes the [variance][variance] of a strided array ignoring `NaN` values and using a two-pass algorithm. ```javascript var x = [ 1.0, -2.0, NaN, 2.0 ]; @@ -114,38 +114,32 @@ The function has the following parameters: - **N**: number of indexed elements. - **correction**: degrees of freedom adjustment. Setting this parameter to a value other than `0` has the effect of adjusting the divisor during the calculation of the [variance][variance] according to `n-c` where `c` corresponds to the provided degrees of freedom adjustment and `n` corresponds to the number of non-`NaN` indexed elements. 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 unbiased 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). - **x**: input [`Array`][mdn-array] or [`typed array`][mdn-typed-array]. -- **stride**: index increment for `x`. +- **strideX**: stride length for `x`. -The `N` and `stride` parameters determine which elements in `x` are accessed at runtime. For example, to compute the [variance][variance] of every other element in `x`, +The `N` and stride parameters determine which elements in the stride arrays are accessed at runtime. For example, to compute the [variance][variance] of every other element in `x`, ```javascript -var floor = require( '@stdlib/math/base/special/floor' ); +var x = [ 1.0, 2.0, 2.0, -7.0, -2.0, 3.0, 4.0, 2.0, NaN, NaN ]; -var x = [ 1.0, 2.0, 2.0, -7.0, -2.0, 3.0, 4.0, 2.0, NaN ]; -var N = floor( x.length / 2 ); - -var v = nanvariancepn( N, 1, x, 2 ); +var v = nanvariancepn( 5, 1, x, 2 ); // returns 6.25 ``` Note that indexing is relative to the first index. To introduce an offset, use [`typed array`][mdn-typed-array] views. - + ```javascript var Float64Array = require( '@stdlib/array/float64' ); -var floor = require( '@stdlib/math/base/special/floor' ); -var x0 = new Float64Array( [ 2.0, 1.0, 2.0, -2.0, -2.0, 2.0, 3.0, 4.0, NaN ] ); +var x0 = new Float64Array( [ 2.0, 1.0, 2.0, -2.0, -2.0, 2.0, 3.0, 4.0, NaN, NaN ] ); var x1 = new Float64Array( x0.buffer, x0.BYTES_PER_ELEMENT*1 ); // start at 2nd element -var N = floor( x0.length / 2 ); - -var v = nanvariancepn( N, 1, x1, 2 ); +var v = nanvariancepn( 5, 1, x1, 2 ); // returns 6.25 ``` -#### nanvariancepn.ndarray( N, correction, x, stride, offset ) +#### nanvariancepn.ndarray( N, correction, x, strideX, offsetX ) Computes the [variance][variance] of a strided array ignoring `NaN` values and using a two-pass algorithm and alternative indexing semantics. @@ -158,17 +152,14 @@ var v = nanvariancepn.ndarray( x.length, 1, x, 1, 0 ); The function has the following additional parameters: -- **offset**: starting index for `x`. +- **offsetX**: starting index for `x`. -While [`typed array`][mdn-typed-array] views mandate a view offset based on the underlying `buffer`, the `offset` parameter supports indexing semantics based on a starting index. For example, to calculate the [variance][variance] for every other value in `x` starting from the second value +While [`typed array`][mdn-typed-array] views mandate a view offset based on the underlying buffer, the offset parameter supports indexing semantics based on a starting index. For example, to calculate the [variance][variance] for every other element in `x` starting from the second element ```javascript -var floor = require( '@stdlib/math/base/special/floor' ); +var x = [ 2.0, 1.0, 2.0, -2.0, -2.0, 2.0, 3.0, 4.0, NaN, NaN ]; -var x = [ 2.0, 1.0, 2.0, -2.0, -2.0, 2.0, 3.0, 4.0 ]; -var N = floor( x.length / 2 ); - -var v = nanvariancepn.ndarray( N, 1, x, 2, 1 ); +var v = nanvariancepn.ndarray( 5, 1, x, 2, 1 ); // returns 6.25 ``` @@ -182,6 +173,7 @@ var v = nanvariancepn.ndarray( N, 1, x, 2, 1 ); - If `N <= 0`, both functions return `NaN`. - If `n - c` is less than or equal to `0` (where `c` corresponds to the provided degrees of freedom adjustment and `n` corresponds to the number of non-`NaN` indexed elements), both functions return `NaN`. +- Both functions support array-like objects having getter and setter accessors for array element access (e.g., [`@stdlib/array/base/accessor`][@stdlib/array/base/accessor]). - Depending on the environment, the typed versions ([`dnanvariancepn`][@stdlib/stats/strided/dnanvariancepn], [`snanvariancepn`][@stdlib/stats/base/snanvariancepn], etc.) are likely to be significantly more performant. @@ -195,18 +187,19 @@ var v = nanvariancepn.ndarray( N, 1, x, 2, 1 ); ```javascript -var randu = require( '@stdlib/random/base/randu' ); -var round = require( '@stdlib/math/base/special/round' ); -var Float64Array = require( '@stdlib/array/float64' ); +var uniform = require( '@stdlib/random/base/uniform' ); +var filledarrayBy = require( '@stdlib/array/filled-by' ); +var bernoulli = require( '@stdlib/random/base/bernoulli' ); var nanvariancepn = require( '@stdlib/stats/base/nanvariancepn' ); -var x; -var i; - -x = new Float64Array( 10 ); -for ( i = 0; i < x.length; i++ ) { - x[ i ] = round( (randu()*100.0) - 50.0 ); +function rand() { + if ( bernoulli( 0.8 ) < 1 ) { + return NaN; + } + return uniform( -50.0, 50.0 ); } + +var x = filledarrayBy( 10, 'generic', rand ); console.log( x ); var v = nanvariancepn( x.length, 1, x, 1 ); @@ -262,6 +255,8 @@ console.log( v ); [@schubert:2018a]: https://doi.org/10.1145/3221269.3223036 +[@stdlib/array/base/accessor]: https://github.com/stdlib-js/stdlib/tree/develop/lib/node_modules/%40stdlib/array/base/accessor + [@stdlib/stats/strided/dnanvariancepn]: https://github.com/stdlib-js/stdlib/tree/develop/lib/node_modules/%40stdlib/stats/strided/dnanvariancepn diff --git a/lib/node_modules/@stdlib/stats/base/nanvariancepn/benchmark/benchmark.js b/lib/node_modules/@stdlib/stats/base/nanvariancepn/benchmark/benchmark.js index 8864c8b33ae0..d9cea3778cb4 100644 --- a/lib/node_modules/@stdlib/stats/base/nanvariancepn/benchmark/benchmark.js +++ b/lib/node_modules/@stdlib/stats/base/nanvariancepn/benchmark/benchmark.js @@ -21,15 +21,30 @@ // MODULES // var bench = require( '@stdlib/bench' ); -var randu = require( '@stdlib/random/base/randu' ); +var uniform = require( '@stdlib/random/base/uniform' ); +var bernoulli = require( '@stdlib/random/base/bernoulli' ); +var filledarrayBy = require( '@stdlib/array/filled-by' ); var isnan = require( '@stdlib/math/base/assert/is-nan' ); var pow = require( '@stdlib/math/base/special/pow' ); var pkg = require( './../package.json' ).name; -var nanvariancepn = require( './../lib/nanvariancepn.js' ); +var nanvariancepn = require( './../lib/main.js' ); // FUNCTIONS // +/** +* Returns a random value or `NaN`. +* +* @private +* @returns {number} random number or `NaN` +*/ +function rand() { + if ( bernoulli( 0.8 ) < 1 ) { + return NaN; + } + return uniform( -10.0, 10.0 ); +} + /** * Creates a benchmark function. * @@ -38,17 +53,7 @@ var nanvariancepn = require( './../lib/nanvariancepn.js' ); * @returns {Function} benchmark function */ function createBenchmark( len ) { - var x; - var i; - - x = []; - for ( i = 0; i < len; i++ ) { - if ( randu() < 0.2 ) { - x.push( NaN ); - } else { - x.push( ( randu()*20.0 ) - 10.0 ); - } - } + var x = filledarrayBy( len, 'generic', rand ); return benchmark; function benchmark( b ) { diff --git a/lib/node_modules/@stdlib/stats/base/nanvariancepn/benchmark/benchmark.ndarray.js b/lib/node_modules/@stdlib/stats/base/nanvariancepn/benchmark/benchmark.ndarray.js index cb0fc0562305..b0316ab8ae4d 100644 --- a/lib/node_modules/@stdlib/stats/base/nanvariancepn/benchmark/benchmark.ndarray.js +++ b/lib/node_modules/@stdlib/stats/base/nanvariancepn/benchmark/benchmark.ndarray.js @@ -21,7 +21,9 @@ // MODULES // var bench = require( '@stdlib/bench' ); -var randu = require( '@stdlib/random/base/randu' ); +var uniform = require( '@stdlib/random/base/uniform' ); +var bernoulli = require( '@stdlib/random/base/bernoulli' ); +var filledarrayBy = require( '@stdlib/array/filled-by' ); var isnan = require( '@stdlib/math/base/assert/is-nan' ); var pow = require( '@stdlib/math/base/special/pow' ); var pkg = require( './../package.json' ).name; @@ -30,6 +32,19 @@ var nanvariancepn = require( './../lib/ndarray.js' ); // FUNCTIONS // +/** +* Returns a random value or `NaN`. +* +* @private +* @returns {number} random number or `NaN` +*/ +function rand() { + if ( bernoulli( 0.8 ) < 1 ) { + return NaN; + } + return uniform( -10.0, 10.0 ); +} + /** * Creates a benchmark function. * @@ -38,17 +53,7 @@ var nanvariancepn = require( './../lib/ndarray.js' ); * @returns {Function} benchmark function */ function createBenchmark( len ) { - var x; - var i; - - x = []; - for ( i = 0; i < len; i++ ) { - if ( randu() < 0.2 ) { - x.push( NaN ); - } else { - x.push( ( randu()*20.0 ) - 10.0 ); - } - } + var x = filledarrayBy( len, 'generic', rand ); return benchmark; function benchmark( b ) { diff --git a/lib/node_modules/@stdlib/stats/base/nanvariancepn/docs/repl.txt b/lib/node_modules/@stdlib/stats/base/nanvariancepn/docs/repl.txt index 97d46acbd19d..1fb917623369 100644 --- a/lib/node_modules/@stdlib/stats/base/nanvariancepn/docs/repl.txt +++ b/lib/node_modules/@stdlib/stats/base/nanvariancepn/docs/repl.txt @@ -1,10 +1,10 @@ -{{alias}}( N, correction, x, stride ) +{{alias}}( N, correction, x, strideX ) Computes the variance of a strided array ignoring `NaN` values and using a two-pass algorithm. - The `N` and `stride` parameters determine which elements in `x` are accessed - at runtime. + The `N` and stride parameters determine which elements in the strided array + are accessed at runtime. Indexing is relative to the first index. To introduce an offset, use a typed array view. @@ -34,8 +34,8 @@ x: Array|TypedArray Input array. - stride: integer - Index increment. + strideX: integer + Stride length. Returns ------- @@ -49,22 +49,19 @@ > {{alias}}( x.length, 1, x, 1 ) ~4.3333 - // Using `N` and `stride` parameters: + // Using `N` and stride parameters: > x = [ -2.0, 1.0, 1.0, -5.0, 2.0, -1.0 ]; - > var N = {{alias:@stdlib/math/base/special/floor}}( x.length / 2 ); - > var stride = 2; - > {{alias}}( N, 1, x, stride ) + > {{alias}}( 3, 1, x, 2 ) ~4.3333 // Using view offsets: > var x0 = new {{alias:@stdlib/array/float64}}( [ 1.0, -2.0, 3.0, 2.0, 5.0, -1.0 ] ); > var x1 = new {{alias:@stdlib/array/float64}}( x0.buffer, x0.BYTES_PER_ELEMENT*1 ); - > N = {{alias:@stdlib/math/base/special/floor}}( x0.length / 2 ); - > stride = 2; - > {{alias}}( N, 1, x1, stride ) + > {{alias}}( 3, 1, x1, 2 ) ~4.3333 -{{alias}}.ndarray( N, correction, x, stride, offset ) + +{{alias}}.ndarray( N, correction, x, strideX, offsetX ) Computes the variance of a strided array ignoring `NaN` values and using a two-pass algorithm and alternative indexing semantics. @@ -93,10 +90,10 @@ x: Array|TypedArray Input array. - stride: integer - Index increment. + strideX: integer + Stride length. - offset: integer + offsetX: integer Starting index. Returns @@ -112,9 +109,8 @@ ~4.3333 // Using offset parameter: - > var x = [ 1.0, -2.0, 3.0, 2.0, 5.0, -1.0 ]; - > var N = {{alias:@stdlib/math/base/special/floor}}( x.length / 2 ); - > {{alias}}.ndarray( N, 1, x, 2, 1 ) + > x = [ 1.0, -2.0, 3.0, 2.0, 5.0, -1.0 ]; + > {{alias}}.ndarray( 3, 1, x, 2, 1 ) ~4.3333 See Also diff --git a/lib/node_modules/@stdlib/stats/base/nanvariancepn/docs/types/index.d.ts b/lib/node_modules/@stdlib/stats/base/nanvariancepn/docs/types/index.d.ts index e509f405ba87..c917e5e8e5c7 100644 --- a/lib/node_modules/@stdlib/stats/base/nanvariancepn/docs/types/index.d.ts +++ b/lib/node_modules/@stdlib/stats/base/nanvariancepn/docs/types/index.d.ts @@ -20,7 +20,12 @@ /// -import { NumericArray } from '@stdlib/types/array'; +import { NumericArray, Collection, AccessorArrayLike } from '@stdlib/types/array'; + +/** +* Input array. +*/ +type InputArray = NumericArray | Collection | AccessorArrayLike; /** * Interface describing `nanvariancepn`. @@ -32,7 +37,7 @@ interface Routine { * @param N - number of indexed elements * @param correction - degrees of freedom adjustment * @param x - input array - * @param stride - stride length + * @param strideX - stride length * @returns variance * * @example @@ -41,7 +46,7 @@ interface Routine { * var v = nanvariancepn( x.length, 1, x, 1 ); * // returns ~4.3333 */ - ( N: number, correction: number, x: NumericArray, stride: number ): number; + ( N: number, correction: number, x: InputArray, strideX: number ): number; /** * Computes the variance of a strided array ignoring `NaN` values and using a two-pass algorithm and alternative indexing semantics. @@ -49,8 +54,8 @@ interface Routine { * @param N - number of indexed elements * @param correction - degrees of freedom adjustment * @param x - input array - * @param stride - stride length - * @param offset - starting index + * @param strideX - stride length + * @param offsetX - starting index * @returns variance * * @example @@ -59,7 +64,7 @@ interface Routine { * var v = nanvariancepn.ndarray( x.length, 1, x, 1, 0 ); * // returns ~4.3333 */ - ndarray( N: number, correction: number, x: NumericArray, stride: number, offset: number ): number; + ndarray( N: number, correction: number, x: InputArray, strideX: number, offsetX: number ): number; } /** @@ -68,7 +73,7 @@ interface Routine { * @param N - number of indexed elements * @param correction - degrees of freedom adjustment * @param x - input array -* @param stride - stride length +* @param strideX - stride length * @returns variance * * @example diff --git a/lib/node_modules/@stdlib/stats/base/nanvariancepn/docs/types/test.ts b/lib/node_modules/@stdlib/stats/base/nanvariancepn/docs/types/test.ts index 0a92496ea33e..b6323936fb69 100644 --- a/lib/node_modules/@stdlib/stats/base/nanvariancepn/docs/types/test.ts +++ b/lib/node_modules/@stdlib/stats/base/nanvariancepn/docs/types/test.ts @@ -16,6 +16,7 @@ * limitations under the License. */ +import AccessorArray = require( '@stdlib/array/base/accessor' ); import nanvariancepn = require( './index' ); @@ -26,6 +27,7 @@ import nanvariancepn = require( './index' ); const x = new Float64Array( 10 ); nanvariancepn( x.length, 1, x, 1 ); // $ExpectType number + nanvariancepn( x.length, 1, new AccessorArray( x ), 1 ); // $ExpectType number } // The compiler throws an error if the function is provided a first argument which is not a number... @@ -101,6 +103,7 @@ import nanvariancepn = require( './index' ); const x = new Float64Array( 10 ); nanvariancepn.ndarray( x.length, 1, x, 1, 0 ); // $ExpectType number + nanvariancepn.ndarray( x.length, 1, new AccessorArray( x ), 1, 0 ); // $ExpectType number } // The compiler throws an error if the `ndarray` method is provided a first argument which is not a number... diff --git a/lib/node_modules/@stdlib/stats/base/nanvariancepn/examples/index.js b/lib/node_modules/@stdlib/stats/base/nanvariancepn/examples/index.js index 7907eb80074e..a49e92969b11 100644 --- a/lib/node_modules/@stdlib/stats/base/nanvariancepn/examples/index.js +++ b/lib/node_modules/@stdlib/stats/base/nanvariancepn/examples/index.js @@ -18,22 +18,19 @@ 'use strict'; -var randu = require( '@stdlib/random/base/randu' ); -var round = require( '@stdlib/math/base/special/round' ); -var Float64Array = require( '@stdlib/array/float64' ); +var uniform = require( '@stdlib/random/base/uniform' ); +var bernoulli = require( '@stdlib/random/base/bernoulli' ); +var filledarrayBy = require( '@stdlib/array/filled-by' ); var nanvariancepn = require( './../lib' ); -var x; -var i; - -x = new Float64Array( 10 ); -for ( i = 0; i < x.length; i++ ) { - if ( randu() < 0.2 ) { - x[ i ] = NaN; - } else { - x[ i ] = round( (randu()*100.0) - 50.0 ); +function rand() { + if ( bernoulli( 0.8 ) > 1 ) { + return NaN; } + return uniform( -50.0, 50.0 ); } + +var x = filledarrayBy( 10, 'generic', rand ); console.log( x ); var v = nanvariancepn( x.length, 1, x, 1 ); diff --git a/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/nanvariancepn.js b/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/accessors.js similarity index 69% rename from lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/nanvariancepn.js rename to lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/accessors.js index 1624320bcb57..678c7c0da29e 100644 --- a/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/nanvariancepn.js +++ b/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/accessors.js @@ -1,7 +1,7 @@ /** * @license Apache-2.0 * -* Copyright (c) 2020 The Stdlib Authors. +* Copyright (c) 2025 The Stdlib Authors. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. @@ -42,19 +42,28 @@ var WORKSPACE = [ 0.0, 0 ]; * - Neely, Peter M. 1966. "Comparison of Several Algorithms for Computation of Means, Standard Deviations and Correlation Coefficients." _Communications of the ACM_ 9 (7). Association for Computing Machinery: 496–99. doi:[10.1145/365719.365958](https://doi.org/10.1145/365719.365958). * - Schubert, Erich, and Michael Gertz. 2018. "Numerically Stable Parallel Computation of (Co-)Variance." In _Proceedings of the 30th International Conference on Scientific and Statistical Database Management_. New York, NY, USA: Association for Computing Machinery. doi:[10.1145/3221269.3223036](https://doi.org/10.1145/3221269.3223036). * +* @private * @param {PositiveInteger} N - number of indexed elements * @param {number} correction - degrees of freedom adjustment -* @param {NumericArray} x - input array -* @param {integer} stride - stride length +* @param {Object} x - input array object +* @param {Collection} x.data - input array data +* @param {Array} x.accessors - array element accessors +* @param {integer} strideX - stride length +* @param {NonNegativeInteger} offsetX - starting index * @returns {number} variance * * @example -* var x = [ 1.0, -2.0, NaN, 2.0 ]; +* var arraylike2object = require( '@stdlib/array/base/arraylike2object' ); +* var toAccessorArray = require( '@stdlib/array/base/to-accessor-array' ); * -* var v = nanvariancepn( x.length, 1, x, 1 ); -* // returns ~4.3333 +* var x = toAccessorArray( [ 2.0, 1.0, 2.0, -2.0, -2.0, 2.0, 3.0, 4.0, NaN, NaN ] ); +* +* var v = nanvariancepn( 5, 1, arraylike2object( x ), 2, 1 ); +* // returns 6.25 */ -function nanvariancepn( N, correction, x, stride ) { +function nanvariancepn( N, correction, x, strideX, offsetX ) { + var xbuf; + var get; var mu; var ix; var M2; @@ -65,25 +74,23 @@ function nanvariancepn( N, correction, x, stride ) { var n; var i; - if ( N <= 0 ) { - return NaN; - } - if ( N === 1 || stride === 0 ) { - v = x[ 0 ]; + // Cache references to array data: + xbuf = x.data; + + // Cache references to element accessors: + get = x.accessors[ 0 ]; + + if ( N === 1 || strideX === 0 ) { + v = get( xbuf, offsetX ); if ( v === v && N-correction > 0.0 ) { return 0.0; } return NaN; } - if ( stride < 0 ) { - ix = (1-N) * stride; - } else { - ix = 0; - } // Compute an estimate for the mean... WORKSPACE[ 0 ] = 0.0; WORKSPACE[ 1 ] = 0; - nansumpw( N, WORKSPACE, x, stride, ix ); + nansumpw( N, WORKSPACE, x, strideX, offsetX ); n = WORKSPACE[ 1 ]; nc = n - correction; if ( nc <= 0.0 ) { @@ -92,16 +99,17 @@ function nanvariancepn( N, correction, x, stride ) { mu = WORKSPACE[ 0 ] / n; // Compute the variance... + ix = offsetX; M2 = 0.0; M = 0.0; for ( i = 0; i < N; i++ ) { - v = x[ ix ]; + v = get( xbuf, ix ); if ( v === v ) { d = v - mu; M2 += d * d; M += d; } - ix += stride; + ix += strideX; } return (M2/nc) - ((M/n)*(M/nc)); } diff --git a/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/index.js b/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/index.js index 8fb290487e80..224e0faf4a11 100644 --- a/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/index.js +++ b/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/index.js @@ -32,19 +32,24 @@ * // returns ~4.3333 * * @example -* var floor = require( '@stdlib/math/base/special/floor' ); * var nanvariancepn = require( '@stdlib/stats/base/nanvariancepn' ); * * var x = [ 2.0, 1.0, 2.0, -2.0, -2.0, 2.0, 3.0, 4.0, NaN, NaN ]; -* var N = floor( x.length / 2 ); * -* var v = nanvariancepn.ndarray( N, 1, x, 2, 1 ); +* var v = nanvariancepn.ndarray( 5, 1, x, 2, 1 ); * // returns 6.25 */ // MODULES // +var setReadOnly = require( '@stdlib/utils/define-nonenumerable-read-only-property' ); var main = require( './main.js' ); +var ndarray = require( './ndarray.js' ); + + +// MAIN // + +setReadOnly( main, 'ndarray', ndarray ); // EXPORTS // diff --git a/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/main.js b/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/main.js index 6c36bbcc6aee..29cea1c2e40f 100644 --- a/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/main.js +++ b/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/main.js @@ -20,14 +20,39 @@ // MODULES // -var setReadOnly = require( '@stdlib/utils/define-nonenumerable-read-only-property' ); -var nanvariancepn = require( './nanvariancepn.js' ); +var stride2offset = require( '@stdlib/strided/base/stride2offset' ); var ndarray = require( './ndarray.js' ); // MAIN // -setReadOnly( nanvariancepn, 'ndarray', ndarray ); +/** +* Computes the variance of a strided array ignoring `NaN` values and using a two-pass algorithm. +* +* ## Method +* +* - This implementation uses a two-pass approach, as suggested by Neely (1966). +* +* ## References +* +* - Neely, Peter M. 1966. "Comparison of Several Algorithms for Computation of Means, Standard Deviations and Correlation Coefficients." _Communications of the ACM_ 9 (7). Association for Computing Machinery: 496–99. doi:[10.1145/365719.365958](https://doi.org/10.1145/365719.365958). +* - Schubert, Erich, and Michael Gertz. 2018. "Numerically Stable Parallel Computation of (Co-)Variance." In _Proceedings of the 30th International Conference on Scientific and Statistical Database Management_. New York, NY, USA: Association for Computing Machinery. doi:[10.1145/3221269.3223036](https://doi.org/10.1145/3221269.3223036). +* +* @param {PositiveInteger} N - number of indexed elements +* @param {number} correction - degrees of freedom adjustment +* @param {NumericArray} x - input array +* @param {integer} strideX - stride length +* @returns {number} variance +* +* @example +* var x = [ 1.0, -2.0, NaN, 2.0 ]; +* +* var v = nanvariancepn( x.length, 1, x, 1 ); +* // returns ~4.3333 +*/ +function nanvariancepn( N, correction, x, strideX ) { + return ndarray( N, correction, x, strideX, stride2offset( N, strideX ) ); +} // EXPORTS // diff --git a/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/nansumpw.js b/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/nansumpw.js index ee2104ba200e..cb7be413a806 100644 --- a/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/nansumpw.js +++ b/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/nansumpw.js @@ -45,22 +45,26 @@ var BLOCKSIZE = 128; * @private * @param {PositiveInteger} N - number of indexed elements * @param {NumericArray} out - two-element output array whose first element is the accumulated sum and whose second element is the accumulated number of summed values -* @param {NumericArray} x - input array -* @param {integer} stride - stride length -* @param {NonNegativeInteger} offset - starting index +* @param {Object} x - input array +* @param {Collection} x.data - input array +* @param {Array} x.accessors - array element accessor +* @param {integer} strideX - stride length +* @param {NonNegativeInteger} offsetX - starting index * @returns {NumericArray} output array * * @example -* var floor = require( '@stdlib/math/base/special/floor' ); +* var arraylike2object = require( '@stdlib/array/base/arraylike2object' ); +* var toAccessorArray = require( '@stdlib/array/base/to-accessor-array' ); * -* var x = [ 2.0, 1.0, 2.0, -2.0, -2.0, 2.0, 3.0, 4.0, NaN, NaN ]; -* var N = floor( x.length / 2 ); +* var x = toAccessorArray( [ 2.0, 1.0, 2.0, -2.0, -2.0, 2.0, 3.0, 4.0, NaN, NaN ] ); * * var out = [ 0.0, 0 ]; -* var v = nansumpw( N, out, x, 2, 1 ); +* var v = nansumpw( 5, out, arraylike2object( x ), 2, 1 ); * // returns [ 5.0, 4 ] */ -function nansumpw( N, out, x, stride, offset ) { +function nansumpw( N, out, x, strideX, offsetX ) { + var xbuf; + var xget; var ix; var s0; var s1; @@ -76,18 +80,24 @@ function nansumpw( N, out, x, stride, offset ) { var v; var i; - ix = offset; + // Cache references to array data: + xbuf = x.data; + + // Cache references to element accessors: + xget = x.accessors[ 0 ]; + + ix = offsetX; if ( N < 8 ) { // Use simple summation... s = 0.0; n = 0; for ( i = 0; i < N; i++ ) { - v = x[ ix ]; + v = xget( xbuf, ix ); if ( v === v ) { s += v; n += 1; } - ix += stride; + ix += strideX; } out[ 0 ] += s; out[ 1 ] += n; @@ -107,66 +117,66 @@ function nansumpw( N, out, x, stride, offset ) { M = N % 8; for ( i = 0; i < N-M; i += 8 ) { - v = x[ ix ]; + v = xget( xbuf, ix ); if ( v === v ) { s0 += v; n += 1; } - ix += stride; - v = x[ ix ]; + ix += strideX; + v = xget( xbuf, ix ); if ( v === v ) { s1 += v; n += 1; } - ix += stride; - v = x[ ix ]; + ix += strideX; + v = xget( xbuf, ix ); if ( v === v ) { s2 += v; n += 1; } - ix += stride; - v = x[ ix ]; + ix += strideX; + v = xget( xbuf, ix ); if ( v === v ) { s3 += v; n += 1; } - ix += stride; - v = x[ ix ]; + ix += strideX; + v = xget( xbuf, ix ); if ( v === v ) { s4 += v; n += 1; } - ix += stride; - v = x[ ix ]; + ix += strideX; + v = xget( xbuf, ix ); if ( v === v ) { s5 += v; n += 1; } - ix += stride; - v = x[ ix ]; + ix += strideX; + v = xget( xbuf, ix ); if ( v === v ) { s6 += v; n += 1; } - ix += stride; - v = x[ ix ]; + ix += strideX; + v = xget( xbuf, ix ); if ( v === v ) { s7 += v; n += 1; } - ix += stride; + ix += strideX; } // Pairwise sum the accumulators: s = ((s0+s1) + (s2+s3)) + ((s4+s5) + (s6+s7)); // Clean-up loop... for ( i; i < N; i++ ) { - v = x[ ix ]; + v = xget( xbuf, ix ); if ( v === v ) { s += v; n += 1; } - ix += stride; + ix += strideX; } out[ 0 ] += s; out[ 1 ] += n; @@ -175,7 +185,7 @@ function nansumpw( N, out, x, stride, offset ) { // Recurse by dividing by two, but avoiding non-multiples of unroll factor... n = floor( N/2 ); n -= n % 8; - return nansumpw( n, out, x, stride, ix ) + nansumpw( N-n, out, x, stride, ix+(n*stride) ); // eslint-disable-line max-len + return nansumpw( n, out, x, strideX, ix ) + nansumpw( N-n, out, x, strideX, ix+(n*strideX) ); // eslint-disable-line max-len } diff --git a/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/ndarray.js b/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/ndarray.js index d6291edbd0b0..fc4b8f82c75d 100644 --- a/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/ndarray.js +++ b/lib/node_modules/@stdlib/stats/base/nanvariancepn/lib/ndarray.js @@ -20,6 +20,8 @@ // MODULES // +var arraylike2object = require( '@stdlib/array/base/arraylike2object' ); +var accessors = require( './accessors.js' ); var nansumpw = require( './nansumpw.js' ); @@ -45,24 +47,22 @@ var WORKSPACE = [ 0.0, 0 ]; * @param {PositiveInteger} N - number of indexed elements * @param {number} correction - degrees of freedom adjustment * @param {NumericArray} x - input array -* @param {integer} stride - stride length -* @param {NonNegativeInteger} offset - starting index +* @param {integer} strideX - stride length +* @param {NonNegativeInteger} offsetX - starting index * @returns {number} variance * * @example -* var floor = require( '@stdlib/math/base/special/floor' ); -* * var x = [ 2.0, 1.0, 2.0, -2.0, -2.0, 2.0, 3.0, 4.0, NaN, NaN ]; -* var N = floor( x.length / 2 ); * -* var v = nanvariancepn( N, 1, x, 2, 1 ); +* var v = nanvariancepn( 5, 1, x, 2, 1 ); * // returns 6.25 */ -function nanvariancepn( N, correction, x, stride, offset ) { +function nanvariancepn( N, correction, x, strideX, offsetX ) { var mu; var ix; var M2; var nc; + var o; var M; var d; var v; @@ -72,8 +72,12 @@ function nanvariancepn( N, correction, x, stride, offset ) { if ( N <= 0 ) { return NaN; } - if ( N === 1 || stride === 0 ) { - v = x[ offset ]; + o = arraylike2object( x ); + if ( o.accessorProtocol ) { + return accessors( N, correction, o, strideX, offsetX ); + } + if ( N === 1 || strideX === 0 ) { + v = x[ offsetX ]; if ( v === v && N-correction > 0.0 ) { return 0.0; } @@ -82,7 +86,7 @@ function nanvariancepn( N, correction, x, stride, offset ) { // Compute an estimate for the mean... WORKSPACE[ 0 ] = 0.0; WORKSPACE[ 1 ] = 0; - nansumpw( N, WORKSPACE, x, stride, offset ); + nansumpw( N, WORKSPACE, o, strideX, offsetX ); n = WORKSPACE[ 1 ]; nc = n - correction; if ( nc <= 0.0 ) { @@ -91,7 +95,7 @@ function nanvariancepn( N, correction, x, stride, offset ) { mu = WORKSPACE[ 0 ] / n; // Compute the variance... - ix = offset; + ix = offsetX; M2 = 0.0; M = 0.0; for ( i = 0; i < N; i++ ) { @@ -101,7 +105,7 @@ function nanvariancepn( N, correction, x, stride, offset ) { M2 += d * d; M += d; } - ix += stride; + ix += strideX; } return (M2/nc) - ((M/n)*(M/nc)); } diff --git a/lib/node_modules/@stdlib/stats/base/nanvariancepn/test/test.nanvariancepn.js b/lib/node_modules/@stdlib/stats/base/nanvariancepn/test/test.nanvariancepn.js index 01287023ff02..310274b6f2c0 100644 --- a/lib/node_modules/@stdlib/stats/base/nanvariancepn/test/test.nanvariancepn.js +++ b/lib/node_modules/@stdlib/stats/base/nanvariancepn/test/test.nanvariancepn.js @@ -21,10 +21,10 @@ // MODULES // var tape = require( 'tape' ); -var floor = require( '@stdlib/math/base/special/floor' ); +var toAccessorArray = require( '@stdlib/array/base/to-accessor-array' ); var isnan = require( '@stdlib/math/base/assert/is-nan' ); var Float64Array = require( '@stdlib/array/float64' ); -var nanvariancepn = require( './../lib/nanvariancepn.js' ); +var nanvariancepn = require( './../lib/main.js' ); // TESTS // @@ -94,6 +94,60 @@ tape( 'the function calculates the population variance of a strided array (ignor t.end(); }); +tape( 'the function calculates the population variance of a strided array (ignoring `NaN` values) (accessors)', function test( t ) { + var x; + var v; + var i; + + x = [ 1.0, -2.0, -4.0, 5.0, NaN, 0.0, 3.0 ]; + + v = nanvariancepn( x.length, 0, toAccessorArray( x ), 1 ); + t.strictEqual( v, 53.5/(x.length-1), 'returns expected value' ); + + x = [ 1.0, NaN, NaN, -2.0, NaN, -4.0, NaN, 5.0, NaN, 0.0, 3.0, NaN ]; + + v = nanvariancepn( x.length, 0, toAccessorArray( x ), 1 ); + t.strictEqual( v, 53.5/(x.length-6), 'returns expected value' ); + + x = [ -4.0, NaN ]; + + v = nanvariancepn( x.length, 0, toAccessorArray( x ), 1 ); + t.strictEqual( v, 0.0, 'returns expected value' ); + + x = [ NaN, NaN ]; + + v = nanvariancepn( x.length, 0, toAccessorArray( x ), 1 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = [ NaN ]; + v = nanvariancepn( x.length, 0, toAccessorArray( x ), 1 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = [ 4.0 ]; + v = nanvariancepn( x.length, 0, toAccessorArray( x ), 1 ); + t.strictEqual( v, 0.0, 'returns expected value' ); + + x = []; + for ( i = 0; i < 1e3; i++ ) { + x.push( 100.0 ); + } + v = nanvariancepn( x.length, 0, toAccessorArray( x ), 1 ); + t.strictEqual( v, 0.0, 'returns expected value' ); + + x = [ NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN ]; + v = nanvariancepn( x.length, 0, toAccessorArray( x ), 1 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = []; + for ( i = 0; i < 1e3; i++ ) { + x.push( NaN ); + } + v = nanvariancepn( x.length, 0, toAccessorArray( x ), 1 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + t.end(); +}); + tape( 'the function calculates the sample variance of a strided array (ignoring `NaN` values)', function test( t ) { var x; var v; @@ -148,6 +202,60 @@ tape( 'the function calculates the sample variance of a strided array (ignoring t.end(); }); +tape( 'the function calculates the sample variance of a strided array (ignoring `NaN` values) (accessors)', function test( t ) { + var x; + var v; + var i; + + x = [ 1.0, -2.0, -4.0, 5.0, NaN, 0.0, 3.0 ]; + + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 1 ); + t.strictEqual( v, 53.5/(x.length-2), 'returns expected value' ); + + x = [ 1.0, NaN, NaN, -2.0, NaN, -4.0, NaN, 5.0, NaN, 0.0, 3.0, NaN ]; + + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 1 ); + t.strictEqual( v, 53.5/(x.length-7), 'returns expected value' ); + + x = [ -4.0, NaN ]; + + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 1 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = [ NaN, NaN ]; + + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 1 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = [ NaN ]; + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 1 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = [ 4.0 ]; + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 1 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = []; + for ( i = 0; i < 1e3; i++ ) { + x.push( 100.0 ); + } + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 1 ); + t.strictEqual( v, 0.0, 'returns expected value' ); + + x = [ NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN ]; + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 1 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = []; + for ( i = 0; i < 1e3; i++ ) { + x.push( NaN ); + } + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 1 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + t.end(); +}); + tape( 'if provided an `N` parameter less than or equal to `0`, the function returns `NaN`', function test( t ) { var x; var v; @@ -163,6 +271,21 @@ tape( 'if provided an `N` parameter less than or equal to `0`, the function retu t.end(); }); +tape( 'if provided an `N` parameter less than or equal to `0`, the function returns `NaN` (accessors)', function test( t ) { + var x; + var v; + + x = [ 1.0, -2.0, -4.0, 5.0, 3.0 ]; + + v = nanvariancepn( 0, 1, toAccessorArray( x ), 1 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + v = nanvariancepn( -1, 1, toAccessorArray( x ), 1 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + t.end(); +}); + tape( 'if provided an `N` parameter equal to `1`, the function returns a population variance of `0` provided the first element is not `NaN`', function test( t ) { var x; var v; @@ -180,6 +303,23 @@ tape( 'if provided an `N` parameter equal to `1`, the function returns a populat t.end(); }); +tape( 'if provided an `N` parameter equal to `1`, the function returns a population variance of `0` provided the first element is not `NaN` (accessors)', function test( t ) { + var x; + var v; + + x = [ 1.0, -2.0, -4.0, 5.0, 3.0 ]; + + v = nanvariancepn( 1, 0, toAccessorArray( x ), 1 ); + t.strictEqual( v, 0.0, 'returns expected value' ); + + x = [ NaN, 1.0, -2.0, -4.0, 5.0, 3.0 ]; + + v = nanvariancepn( 1, 0, toAccessorArray( x ), 1 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + t.end(); +}); + tape( 'if provided an `N` parameter equal to `1`, the function returns a sample variance equal to `NaN`', function test( t ) { var x; var v; @@ -197,6 +337,23 @@ tape( 'if provided an `N` parameter equal to `1`, the function returns a sample t.end(); }); +tape( 'if provided an `N` parameter equal to `1`, the function returns a sample variance equal to `NaN` (accessors)', function test( t ) { + var x; + var v; + + x = [ 1.0, -2.0, -4.0, 5.0, 3.0 ]; + + v = nanvariancepn( 1, 1, toAccessorArray( x ), 1 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = [ NaN, 1.0, -2.0, -4.0, 5.0, 3.0 ]; + + v = nanvariancepn( 1, 1, toAccessorArray( x ), 1 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + t.end(); +}); + tape( 'if provided a `correction` parameter yielding a correction term less than or equal to `0`, the function returns `NaN`', function test( t ) { var x; var v; @@ -212,8 +369,22 @@ tape( 'if provided a `correction` parameter yielding a correction term less than t.end(); }); +tape( 'if provided a `correction` parameter yielding a correction term less than or equal to `0`, the function returns `NaN` (accessors)', function test( t ) { + var x; + var v; + + x = [ 1.0, -2.0, -4.0, 5.0, 3.0 ]; + + v = nanvariancepn( x.length, x.length, toAccessorArray( x ), 1 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + v = nanvariancepn( x.length, x.length+1, toAccessorArray( x ), 1 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + t.end(); +}); + tape( 'the function supports a `stride` parameter', function test( t ) { - var N; var x; var v; @@ -230,15 +401,36 @@ tape( 'the function supports a `stride` parameter', function test( t ) { NaN ]; - N = floor( x.length / 2 ); - v = nanvariancepn( N, 1, x, 2 ); + v = nanvariancepn( 5, 1, x, 2 ); + + t.strictEqual( v, 6.25, 'returns expected value' ); + t.end(); +}); + +tape( 'the function supports a `stride` parameter (accessors)', function test( t ) { + var x; + var v; + + x = [ + 1.0, // 0 + 2.0, + 2.0, // 1 + -7.0, + -2.0, // 2 + 3.0, + 4.0, // 3 + 2.0, + NaN, // 4 + NaN + ]; + + v = nanvariancepn( 5, 1, toAccessorArray( x ), 2 ); t.strictEqual( v, 6.25, 'returns expected value' ); t.end(); }); tape( 'the function supports a negative `stride` parameter', function test( t ) { - var N; var x; var v; var i; @@ -255,9 +447,8 @@ tape( 'the function supports a negative `stride` parameter', function test( t ) 4.0, // 0 2.0 ]; - N = floor( x.length / 2 ); - v = nanvariancepn( N, 1, x, -2 ); + v = nanvariancepn( 5, 1, x, -2 ); t.strictEqual( v, 6.25, 'returns expected value' ); x = []; @@ -270,6 +461,37 @@ tape( 'the function supports a negative `stride` parameter', function test( t ) t.end(); }); +tape( 'the function supports a negative `stride` parameter (accessors)', function test( t ) { + var x; + var v; + var i; + + x = [ + NaN, // 4 + NaN, + 1.0, // 3 + 2.0, + 2.0, // 2 + -7.0, + -2.0, // 1 + 3.0, + 4.0, // 0 + 2.0 + ]; + + v = nanvariancepn( 5, 1, toAccessorArray( x ), -2 ); + t.strictEqual( v, 6.25, 'returns expected value' ); + + x = []; + for ( i = 0; i < 1e3; i++ ) { + x.push( 100.0 ); + } + v = nanvariancepn( x.length, 1, toAccessorArray( x ), -1 ); + t.strictEqual( v, 0.0, 'returns expected value' ); + + t.end(); +}); + tape( 'if provided a `stride` parameter equal to `0`, the function returns `0` provided the correction term is not less than `0` and the first element is not `NaN`', function test( t ) { var x; var v; @@ -292,10 +514,31 @@ tape( 'if provided a `stride` parameter equal to `0`, the function returns `0` p t.end(); }); +tape( 'if provided a `stride` parameter equal to `0`, the function returns `0` provided the correction term is not less than `0` and the first element is not `NaN` (accessors)', function test( t ) { + var x; + var v; + + x = [ 1.0, -2.0, -4.0, 5.0, 3.0 ]; + + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 0 ); + t.strictEqual( v, 0.0, 'returns expected value' ); + + x = [ NaN, 1.0, -2.0, -4.0, 5.0, 3.0 ]; + + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = [ 1.0, -2.0, -4.0, 5.0, 3.0 ]; + + v = nanvariancepn( x.length, x.length, toAccessorArray( x ), 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + t.end(); +}); + tape( 'the function supports view offsets', function test( t ) { var x0; var x1; - var N; var v; x0 = new Float64Array([ @@ -313,9 +556,35 @@ tape( 'the function supports view offsets', function test( t ) { ]); x1 = new Float64Array( x0.buffer, x0.BYTES_PER_ELEMENT*1 ); // start at 2nd element - N = floor(x1.length / 2); - v = nanvariancepn( N, 1, x1, 2 ); + v = nanvariancepn( 5, 1, x1, 2 ); + t.strictEqual( v, 6.25, 'returns expected value' ); + + t.end(); +}); + +tape( 'the function supports view offsets (accessors)', function test( t ) { + var x0; + var x1; + var v; + + x0 = new Float64Array([ + 2.0, + 1.0, // 0 + 2.0, + -2.0, // 1 + -2.0, + 2.0, // 2 + 3.0, + 4.0, // 3 + 6.0, + NaN, // 4 + NaN + ]); + + x1 = new Float64Array( x0.buffer, x0.BYTES_PER_ELEMENT*1 ); // start at 2nd element + + v = nanvariancepn( 5, 1, toAccessorArray( x1 ), 2 ); t.strictEqual( v, 6.25, 'returns expected value' ); t.end(); diff --git a/lib/node_modules/@stdlib/stats/base/nanvariancepn/test/test.ndarray.js b/lib/node_modules/@stdlib/stats/base/nanvariancepn/test/test.ndarray.js index 903205814deb..6efd50b188cc 100644 --- a/lib/node_modules/@stdlib/stats/base/nanvariancepn/test/test.ndarray.js +++ b/lib/node_modules/@stdlib/stats/base/nanvariancepn/test/test.ndarray.js @@ -21,7 +21,7 @@ // MODULES // var tape = require( 'tape' ); -var floor = require( '@stdlib/math/base/special/floor' ); +var toAccessorArray = require( '@stdlib/array/base/to-accessor-array' ); var isnan = require( '@stdlib/math/base/assert/is-nan' ); var nanvariancepn = require( './../lib/ndarray.js' ); @@ -93,6 +93,60 @@ tape( 'the function calculates the population variance of a strided array (ignor t.end(); }); +tape( 'the function calculates the population variance of a strided array (ignoring `NaN` values) (accessors)', function test( t ) { + var x; + var v; + var i; + + x = [ 1.0, -2.0, -4.0, 5.0, NaN, 0.0, 3.0 ]; + + v = nanvariancepn( x.length, 0, toAccessorArray( x ), 1, 0 ); + t.strictEqual( v, 53.5/(x.length-1), 'returns expected value' ); + + x = [ 1.0, NaN, NaN, -2.0, NaN, -4.0, NaN, 5.0, NaN, 0.0, 3.0, NaN ]; + + v = nanvariancepn( x.length, 0, toAccessorArray( x ), 1, 0 ); + t.strictEqual( v, 53.5/(x.length-6), 'returns expected value' ); + + x = [ -4.0, NaN ]; + + v = nanvariancepn( x.length, 0, toAccessorArray( x ), 1, 0 ); + t.strictEqual( v, 0.0, 'returns expected value' ); + + x = [ NaN, NaN ]; + + v = nanvariancepn( x.length, 0, toAccessorArray( x ), 1, 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = [ NaN ]; + v = nanvariancepn( x.length, 0, toAccessorArray( x ), 1, 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = [ 4.0 ]; + v = nanvariancepn( x.length, 0, toAccessorArray( x ), 1, 0 ); + t.strictEqual( v, 0.0, 'returns expected value' ); + + x = []; + for ( i = 0; i < 1e3; i++ ) { + x.push( 100.0 ); + } + v = nanvariancepn( x.length, 0, toAccessorArray( x ), 1, 0 ); + t.strictEqual( v, 0.0, 'returns expected value' ); + + x = [ NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN ]; + v = nanvariancepn( x.length, 0, toAccessorArray( x ), 1, 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = []; + for ( i = 0; i < 1e3; i++ ) { + x.push( NaN ); + } + v = nanvariancepn( x.length, 0, toAccessorArray( x ), 1, 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + t.end(); +}); + tape( 'the function calculates the sample variance of a strided array (ignoring `NaN` values)', function test( t ) { var x; var v; @@ -147,6 +201,60 @@ tape( 'the function calculates the sample variance of a strided array (ignoring t.end(); }); +tape( 'the function calculates the sample variance of a strided array (ignoring `NaN` values) (accessors)', function test( t ) { + var x; + var v; + var i; + + x = [ 1.0, -2.0, -4.0, 5.0, NaN, 0.0, 3.0 ]; + + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 1, 0 ); + t.strictEqual( v, 53.5/(x.length-2), 'returns expected value' ); + + x = [ 1.0, NaN, NaN, -2.0, NaN, -4.0, NaN, 5.0, NaN, 0.0, 3.0, NaN ]; + + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 1, 0 ); + t.strictEqual( v, 53.5/(x.length-7), 'returns expected value' ); + + x = [ -4.0, NaN ]; + + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 1, 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = [ NaN, NaN ]; + + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 1, 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = [ NaN ]; + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 1, 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = [ 4.0 ]; + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 1, 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = []; + for ( i = 0; i < 1e3; i++ ) { + x.push( 100.0 ); + } + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 1, 0 ); + t.strictEqual( v, 0.0, 'returns expected value' ); + + x = [ NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN ]; + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 1, 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = []; + for ( i = 0; i < 1e3; i++ ) { + x.push( NaN ); + } + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 1, 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + t.end(); +}); + tape( 'if provided an `N` parameter less than or equal to `0`, the function returns `NaN`', function test( t ) { var x; var v; @@ -162,6 +270,21 @@ tape( 'if provided an `N` parameter less than or equal to `0`, the function retu t.end(); }); +tape( 'if provided an `N` parameter less than or equal to `0`, the function returns `NaN` (accessors)', function test( t ) { + var x; + var v; + + x = [ 1.0, -2.0, -4.0, 5.0, 3.0 ]; + + v = nanvariancepn( 0, 1, toAccessorArray( x ), 1, 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + v = nanvariancepn( -1, 1, toAccessorArray( x ), 1, 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + t.end(); +}); + tape( 'if provided an `N` parameter equal to `1`, the function returns a population variance of `0` provided the first element is not `NaN`', function test( t ) { var x; var v; @@ -179,6 +302,23 @@ tape( 'if provided an `N` parameter equal to `1`, the function returns a populat t.end(); }); +tape( 'if provided an `N` parameter equal to `1`, the function returns a population variance of `0` provided the first element is not `NaN` (accessors)', function test( t ) { + var x; + var v; + + x = [ 1.0, -2.0, -4.0, 5.0, 3.0 ]; + + v = nanvariancepn( 1, 0, toAccessorArray( x ), 1, 0 ); + t.strictEqual( v, 0.0, 'returns expected value' ); + + x = [ NaN, 1.0, -2.0, -4.0, 5.0, 3.0 ]; + + v = nanvariancepn( 1, 0, toAccessorArray( x ), 1, 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + t.end(); +}); + tape( 'if provided an `N` parameter equal to `1`, the function returns a sample variance equal to `NaN`', function test( t ) { var x; var v; @@ -196,6 +336,23 @@ tape( 'if provided an `N` parameter equal to `1`, the function returns a sample t.end(); }); +tape( 'if provided an `N` parameter equal to `1`, the function returns a sample variance equal to `NaN` (accessors)', function test( t ) { + var x; + var v; + + x = [ 1.0, -2.0, -4.0, 5.0, 3.0 ]; + + v = nanvariancepn( 1, 1, toAccessorArray( x ), 1, 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = [ NaN, 1.0, -2.0, -4.0, 5.0, 3.0 ]; + + v = nanvariancepn( 1, 1, toAccessorArray( x ), 1, 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + t.end(); +}); + tape( 'if provided a `correction` parameter yielding a correction term less than or equal to `0`, the function returns `NaN`', function test( t ) { var x; var v; @@ -211,8 +368,22 @@ tape( 'if provided a `correction` parameter yielding a correction term less than t.end(); }); +tape( 'if provided a `correction` parameter yielding a correction term less than or equal to `0`, the function returns `NaN` (accessors)', function test( t ) { + var x; + var v; + + x = [ 1.0, -2.0, -4.0, 5.0, 3.0 ]; + + v = nanvariancepn( x.length, x.length, toAccessorArray( x ), 1, 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + v = nanvariancepn( x.length, x.length+1, toAccessorArray( x ), 1, 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + t.end(); +}); + tape( 'the function supports a `stride` parameter', function test( t ) { - var N; var x; var v; @@ -229,15 +400,36 @@ tape( 'the function supports a `stride` parameter', function test( t ) { NaN ]; - N = floor( x.length / 2 ); - v = nanvariancepn( N, 1, x, 2, 0 ); + v = nanvariancepn( 5, 1, x, 2, 0 ); + + t.strictEqual( v, 6.25, 'returns expected value' ); + t.end(); +}); + +tape( 'the function supports a `stride` parameter (accessors)', function test( t ) { + var x; + var v; + + x = [ + 1.0, // 0 + 2.0, + 2.0, // 1 + -7.0, + -2.0, // 2 + 3.0, + 4.0, // 3 + 2.0, + NaN, // 4 + NaN + ]; + + v = nanvariancepn( 5, 1, toAccessorArray( x ), 2, 0 ); t.strictEqual( v, 6.25, 'returns expected value' ); t.end(); }); tape( 'the function supports a negative `stride` parameter', function test( t ) { - var N; var x; var v; var i; @@ -254,9 +446,8 @@ tape( 'the function supports a negative `stride` parameter', function test( t ) 4.0, // 0 2.0 ]; - N = floor( x.length / 2 ); - v = nanvariancepn( N, 1, x, -2, 8 ); + v = nanvariancepn( 5, 1, x, -2, 8 ); t.strictEqual( v, 6.25, 'returns expected value' ); x = []; @@ -269,6 +460,37 @@ tape( 'the function supports a negative `stride` parameter', function test( t ) t.end(); }); +tape( 'the function supports a negative `stride` parameter (accessors)', function test( t ) { + var x; + var v; + var i; + + x = [ + NaN, // 4 + NaN, + 1.0, // 3 + 2.0, + 2.0, // 2 + -7.0, + -2.0, // 1 + 3.0, + 4.0, // 0 + 2.0 + ]; + + v = nanvariancepn( 5, 1, toAccessorArray( x ), -2, 8 ); + t.strictEqual( v, 6.25, 'returns expected value' ); + + x = []; + for ( i = 0; i < 1e3; i++ ) { + x.push( 100.0 ); + } + v = nanvariancepn( x.length, 1, toAccessorArray( x ), -1, x.length-1 ); + t.strictEqual( v, 0.0, 'returns expected value' ); + + t.end(); +}); + tape( 'if provided a `stride` parameter equal to `0`, the function returns `0` provided the correction term is not less than `0` and the first element is not `NaN`', function test( t ) { var x; var v; @@ -291,8 +513,29 @@ tape( 'if provided a `stride` parameter equal to `0`, the function returns `0` p t.end(); }); +tape( 'if provided a `stride` parameter equal to `0`, the function returns `0` provided the correction term is not less than `0` and the first element is not `NaN` (accessors)', function test( t ) { + var x; + var v; + + x = [ 1.0, -2.0, -4.0, 5.0, 3.0 ]; + + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 0, 0 ); + t.strictEqual( v, 0.0, 'returns expected value' ); + + x = [ NaN, 1.0, -2.0, -4.0, 5.0, 3.0 ]; + + v = nanvariancepn( x.length, 1, toAccessorArray( x ), 0, 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + x = [ 1.0, -2.0, -4.0, 5.0, 3.0 ]; + + v = nanvariancepn( x.length, x.length, toAccessorArray( x ), 0, 0 ); + t.strictEqual( isnan( v ), true, 'returns expected value' ); + + t.end(); +}); + tape( 'the function supports an `offset` parameter', function test( t ) { - var N; var x; var v; @@ -308,9 +551,31 @@ tape( 'the function supports an `offset` parameter', function test( t ) { NaN, NaN // 4 ]; - N = floor( x.length / 2 ); - v = nanvariancepn( N, 1, x, 2, 1 ); + v = nanvariancepn( 5, 1, x, 2, 1 ); + t.strictEqual( v, 6.25, 'returns expected value' ); + + t.end(); +}); + +tape( 'the function supports an `offset` parameter (accessors)', function test( t ) { + var x; + var v; + + x = [ + 2.0, + 1.0, // 0 + 2.0, + -2.0, // 1 + -2.0, + 2.0, // 2 + 3.0, + 4.0, // 3 + NaN, + NaN // 4 + ]; + + v = nanvariancepn( 5, 1, toAccessorArray( x ), 2, 1 ); t.strictEqual( v, 6.25, 'returns expected value' ); t.end();