Difference between revisions of "Fractiles"

Revision as of 01:52, 28 January 2016

Fractiles(x)

This function creates a probability distribution treating the elements of «x» as fractiles (percentiles) with equal probability spacing. The first element of «x» is the zeroth fractile (i.e. minimum), the last element is the 1.0 fractile (100th percentile, i.e. maximum). If «x» has n + 1 elements, the i-th value is the (i/n) fractile. The distribution is piecewise uniform -- that is, it assumes a uniform distribution between each pair of adjacent fractiles. The values of «x» must be numerical and ascending.

Declaration:

Fractiles(x: Array Number Ascending)

This function is considered obsolete. If you have a distribution specified by a set of fractiles (percentiles), it is usually best to use the CumDist funtion, which let's you specify an arbitrary (not just equal probability-spaced) fractiles, and interpolates the density function as linear between specified fractiles (or equivalently, interpolates the cumulative function as quadratic). Fractiles is rarely of value but we retain it in Analytica for compatibility with any older models that may have used it in the past.

@@ Line 2: / Line 2: @@
 [[Category:Doc Status D]] <!-- For Lumina use, do not change -->
-==Fractiles(x: Array Number Ascending)==
+==Fractiles(x)==
+This function creates a probability distribution treating the elements of «x» as fractiles (percentiles) with equal probability spacing. The first element of «x» is the zeroth fractile (i.e. minimum), the last element is the 1.0 fractile (100th percentile, i.e. maximum). If «x» has ''n + 1'' elements, the i-th value is the (i/n) fractile. The distribution is piecewise uniform -- that is, it assumes a [[uniform]] distribution between each pair of adjacent fractiles. The values of «x» must be numerical and ascending.
-This creates a probability distribution treating the elements of x as fractiles (percentiles) with equal probability spacing. The first element of x is the zeroth fractile (i.e. minimum), the last element is the 1.0 fractile (100th percentile, i.e. maximum). If x has n+1 elements, the ith value is the (i/n) fractile. The distribution is piecewise uniform -- that is, it assumes a uniform distribution between each pair of adjacent fractiles. The values of x must be numerical and ascending.
+Declaration:
+:[[Fractiles]](x: Array Number Ascending)
-Fractiles is obsolete. If you have a distribution specified by a set of fractiles (percentiles), it is usually best to use [[CumDist]], which let's you specify an arbitrary (not just equal probability-spaced) fractiles, and interpolates the density function as linear between specified fractiles (or equivalently, interpolates the cumulative function as quadratic). Fractiles is rarely of value. We retain it in Analytica simply for compatibility with any (very old) models that may have used it in the past.
+This function is considered obsolete. If you have a distribution specified by a set of fractiles (percentiles), it is usually best to use the [[CumDist]] funtion, which let's you specify an arbitrary (not just equal probability-spaced) fractiles, and interpolates the density function as linear between specified fractiles (or equivalently, interpolates the cumulative function as quadratic). Fractiles is rarely of value but we retain it in Analytica for compatibility with any older models that may have used it in the past.
+==See Also==
+* [[Smooth_Fractile]]
+* [[CumDist]]
+* [[Distribution Densities Library]]