Discrete Fitted distribution | Vose Software

Discrete Fitted distribution

Format: DiscreteFit({data}, min, max, Increment, FillGaps)

An empirical discrete distribution constructed from a set of values {data} where the possible outcomes all lie at a fixed distance Increment from each other.

By default, the minimum and maximum of the DiscreteFit distribution are the data's lowest and highest values. You can provide optional parameters min and max to construct a distribution with a wider range of possible outcome values than the data.

When FillGaps is set to FALSE (default) a zero probability of outcome is assigned to values within the range that never occur in the data - they our considered "impossible" and the probability mass assigned to each outcome value is just its relative frequency in the {data}. Set FillGaps to TRUE to assign a non-zero (albeit small) possibility mass to outcome values that did not occur in the {data}.

By default the increment between possible output values is 1. You can optionally allow for smaller, non-integer increment values by providing an Increment parameter on [0,1]. For the increment value to be valid, (max-min)/increment must be an integer.

Examples

VoseDiscreteFit({1,2,4},,,FALSE) will have only 1,2 and 4 as possible outcome values, each with equal probability. VoseDiscreteFit({1,2,4},,,TRUE) will return 1,2 and 4 with equal probability, and 3 with smaller probability.

VoseDiscreteFit({1,2},0,3,0.5,TRUE) has possible outcomes {0,0.5,1,1.5,2,2.5,3}.

ModelRisk functions added to Microsoft Excel for this distribution

VoseDiscreteFit generates random values from this distribution for Monte Carlo simulation, or calculates a percentile if used with a U parameter.

VoseDiscreteFitProb returns the probability mass or cumulative distribution function for this distribution.

VoseDiscreteFitProb10 returns the log10 of the probability mass or cumulative distribution function.