Skip to content
/ stdlib Public

Files

This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.

Latest commit

kgrytekgryte
chore: fix lint comments
May 14, 2025
cac2aa9 · May 14, 2025

History

History
403 lines (249 loc) · 16.4 KB

README.md

File metadata and controls

403 lines (249 loc) · 16.4 KB

snanvariancepn

Calculate the variance of a single-precision floating-point strided array ignoring NaN values and using a two-pass algorithm.

The population variance of a finite size population of size N is given by

σ 2 = 1 N ∑ i = 0 N − 1 ( x i − μ ) 2

where the population mean is given by

μ = 1 N ∑ i = 0 N − 1 x i

Often in the analysis of data, the true population 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, the result is biased and yields a biased sample variance. To compute an unbiased sample variance for a sample of size n,

s 2 = 1 n − 1 ∑ i = 0 n − 1 ( x i − x ¯ ) 2

where the sample mean is given by

x ¯ = 1 n ∑ i = 0 n − 1 x i

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.

Usage

var snanvariancepn = require( '@stdlib/stats/base/snanvariancepn' );

snanvariancepn( N, correction, x, strideX )

Computes the variance of a single-precision floating-point strided array ignoring NaN values and using a two-pass algorithm.

var Float32Array = require( '@stdlib/array/float32' );

var x = new Float32Array( [ 1.0, -2.0, NaN, 2.0 ] );

var v = snanvariancepn( x.length, 1, x, 1 );
// returns ~4.3333

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 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 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, 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 Float32Array.
  • strideX: stride length for x.

The N and stride parameters determine which elements in the strided array are accessed at runtime. For example, to compute the variance of every other element in x,

var Float32Array = require( '@stdlib/array/float32' );

var x = new Float32Array( [ 1.0, 2.0, 2.0, -7.0, -2.0, 3.0, 4.0, 2.0, NaN, NaN ] );

var v = snanvariancepn( 5, 1, x, 2 );
// returns 6.25

Note that indexing is relative to the first index. To introduce an offset, use typed array views.

var Float32Array = require( '@stdlib/array/float32' );

var x0 = new Float32Array( [ 2.0, 1.0, 2.0, -2.0, -2.0, 2.0, 3.0, 4.0, NaN, NaN ] );
var x1 = new Float32Array( x0.buffer, x0.BYTES_PER_ELEMENT*1 ); // start at 2nd element

var v = snanvariancepn( 5, 1, x1, 2 );
// returns 6.25

snanvariancepn.ndarray( N, correction, x, strideX, offsetX )

Computes the variance of a single-precision floating-point strided array ignoring NaN values and using a two-pass algorithm and alternative indexing semantics.

var Float32Array = require( '@stdlib/array/float32' );

var x = new Float32Array( [ 1.0, -2.0, NaN, 2.0 ] );

var v = snanvariancepn.ndarray( x.length, 1, x, 1, 0 );
// returns ~4.33333

The function has the following additional parameters:

  • offsetX: starting index for x.

While 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 for every other element in x starting from the second element

var Float32Array = require( '@stdlib/array/float32' );

var x = new Float32Array( [ 2.0, 1.0, 2.0, -2.0, -2.0, 2.0, 3.0, 4.0, NaN, NaN ] );

var v = snanvariancepn.ndarray( 5, 1, x, 2, 1 );
// returns 6.25

Notes

  • 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.

Examples

var uniform = require( '@stdlib/random/base/uniform' );
var filledarrayBy = require( '@stdlib/array/filled-by' );
var bernoulli = require( '@stdlib/random/base/bernoulli' );
var snanvariancepn = require( '@stdlib/stats/base/snanvariancepn' );

function rand() {
    if ( bernoulli( 0.8 ) < 1 ) {
        return NaN;
    }
    return uniform( -50.0, 50.0 );
}

var x = filledarrayBy( 10, 'float32', rand );
console.log( x );

var v = snanvariancepn( x.length, 1, x, 1 );
console.log( v );

C APIs

Usage

#include "stdlib/stats/base/snanvariancepn.h"

stdlib_strided_snanvariancepn( N, correction, *X, strideX )

Computes the variance of a single-precision floating-point strided array ignoring NaN values and using a two-pass algorithm.

const float x[] = { 1.0f, -2.0f, 0.0f/0.0f, 2.0f };

float v = stdlib_strided_snanvariancepn( 4, 1.0f, x, 1 );
// returns ~4.3333f

The function accepts the following arguments:

  • N: [in] CBLAS_INT number of indexed elements.
  • correction: [in] float 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 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 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, 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: [in] float* input array.
  • strideX: [in] CBLAS_INT stride length for X.
float stdlib_strided_snanvariancepn( const CBLAS_INT N, const float correction, const float *X, const CBLAS_INT strideX );

stdlib_strided_snanvariancepn_ndarray( N, correction, *X, strideX, offsetX )

Computes the variance of a single-precision floating-point strided array ignoring NaN values and using a two-pass algorithm and alternative indexing semantics.

const float x[] = { 1.0f, -2.0f, 0.0f/0.0f, 2.0f };

float v = stdlib_strided_snanvariancepn_ndarray( 4, 1.0f, x, 1, 0 );
// returns ~4.3333f

The function accepts the following arguments:

  • N: [in] CBLAS_INT number of indexed elements.
  • correction: [in] float 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 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 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, 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: [in] float* input array.
  • strideX: [in] CBLAS_INT stride length for X.
  • offsetX: [in] CBLAS_INT starting index for X.
float stdlib_strided_snanvariancepn_ndarray( const CBLAS_INT N, const float correction, const float *X, const CBLAS_INT strideX, const CBLAS_INT offsetX );

Examples

#include "stdlib/stats/base/snanvariancepn.h"
#include <stdio.h>

int main( void ) {
    // Create a strided array:
    const float x[] = { 1.0f, 2.0f, 0.0f/0.0f, 3.0f, 0.0f/0.0f, 4.0f, 5.0f, 6.0f, 0.0f/0.0f, 7.0f, 8.0f, 0.0f/0.0f };

    // Specify the number of elements:
    const int N = 6;

    // Specify the stride length:
    const int strideX = 2;

    // Compute the variance:
    float v = stdlib_strided_snanvariancepn( N, 1.0f, x, strideX );

    // Print the result:
    printf( "sample variance: %f\n", v );
}

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.
  • 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.

See Also