Logarithm of Probability Density Function

Raised cosine distribution logarithm of probability density function (PDF).

The probability density function (PDF) for a raised cosine random variable is

f left-parenthesis x semicolon mu comma s right-parenthesis equals StartLayout Enlarged left-brace 1st Row 1st Column StartFraction 1 Over 2 s EndFraction left-bracket 1 plus cosine left-parenthesis StartFraction x minus mu Over s EndFraction pi right-parenthesis right-bracket 2nd Column for mu minus s less-than-or-equal-to x less-than-or-equal-to mu plus s 2nd Row 1st Column 0 2nd Column otherwise EndLayout

where μ is the location parameter and s > 0 is the scale parameter.

Usage

var logpdf = require( '@stdlib/stats/base/dists/cosine/logpdf' );

logpdf( x, mu, s )

Evaluates the logarithm of the probability density function (PDF) for a raised cosine distribution with parameters mu (location parameter) and s (scale parameter).

var y = logpdf( 2.0, 0.0, 3.0 );
// returns ~-2.485

y = logpdf( -1.0, 2.0, 4.0 );
// returns ~-3.307

If provided NaN as any argument, the function returns NaN.

var y = logpdf( NaN, 0.0, 1.0 );
// returns NaN

y = logpdf( 0.0, NaN, 1.0 );
// returns NaN

y = logpdf( 0.0, 0.0, NaN );
// returns NaN

If provided s < 0, the function returns NaN.

var y = logpdf( 2.0, 0.0, -1.0 );
// returns NaN

If provided s = 0, the function evaluates the logarithm of the PDF of a degenerate distribution centered at mu.

var y = logpdf( 2.0, 8.0, 0.0 );
// returns -Infinity

y = logpdf( 8.0, 8.0, 0.0 );
// returns Infinity

logpdf.factory( mu, s )

Returns a function for evaluating the logarithm of the probability density function (PDF) of a raised cosine distribution with parameters mu (location parameter) and s (scale parameter).

var mylogpdf = logpdf.factory( 10.0, 2.0 );

var y = mylogpdf( 10.0 );
// returns ~-0.693

y = mylogpdf( 9.0 );
// returns ~-1.386

Notes

  • In virtually all cases, using the logpdf or logcdf functions is preferable to manually computing the logarithm of the pdf or cdf, respectively, since the latter is prone to overflow and underflow.

Examples

var randu = require( '@stdlib/random/base/randu' );
var logpdf = require( '@stdlib/stats/base/dists/cosine/logpdf' );

var mu;
var s;
var x;
var y;
var i;

for ( i = 0; i < 10; i++ ) {
    x = randu() * 10.0;
    mu = randu() * 10.0;
    s = randu() * 10.0;
    y = logpdf( x, mu, s );
    console.log( 'x: %d, µ: %d, s: %d, ln(f(x;µ,s)): %d', x, mu, s, y );
}
Did you find this page helpful?