SED navigation bar go to SED home page go to Dataplot home page go to NIST home page SED Home Page SED Staff SED Projects SED Products and Publications Search SED Pages
Dataplot Vol 2 Vol 1

YULPDF

Name:
    YULPDF (LET)
Type:
    Library Function
Purpose:
    Compute the Yule probability mass function.
Description:
    The Yule distribution has the following probability mass function:

      p(x,p) = p*p!*x!/(x+p+1)!       p > 0, x = 0, 1, 2, ...
 = p*GAMMA(p+1)*GAMMA(x+1)/GAMMA(x+p+2)

    with p denoting the shape parameter and GAMMA denoting the gamma function (HELP GAMMA for details).

    Dataplot computes the Yule probability density function using the log gamma function. The Yule distribution has increasingly long tails as p goes to zero. Currently, Dataplot limits the Yule pdf function to the case where p >= 0.1.

Syntax:
    LET <y> = YULPDF(<x>,<p>)             <SUBSET/EXCEPT/FOR qualification>
    where <x> is a non-negative integer number, parameter, or variable;
                <p> is a positive number, parameter, or variable that specifies the shape parameter;
                <y> is a variable or a parameter (depending on what <x> is) where the computed Yule pdf value is stored;
    and where the <SUBSET/EXCEPT/FOR qualification> is optional.
Examples:
    LET A = YULPDF(3,1.5)
    LET A = YULPDF(X,P)
    PLOT YULPDF(X,2) FOR X = 0 1 50
Note:
    The Yule is a special case of the Waring distribution. Specifically,

      YULPDF(X,P) = WARPDF(X,P-1,1)
Note:
    The following commands for generating random numbers, probability plots, and goodness of fit statistics for the Yule distribution are available:

      LET P = <value>
      LET Y = YULE RANDOM NUMBERS FOR I = 1 1 N
      YULE PROBABILITY PLOT Y
      YULE CHI-SQUARE GOODNESS OF FIT Y

    The following graphical methods for estimating the shape parameter are available:

      YULE PPCC PLOT Y
      YULE KS PLOT Y

    Our experience indicates that the KS PLOT is more effective than the PPCC PLOT for estimating the shape parameter.

    The follwing command

      YULE MAXIMUM LIKELIHOOD Y

    can be used to generate the following estimates of the shape parameter:

    1. The method of moments:

        phat = (xbar + 1)/xbar

      with xbar denoting the sample mean. Note that this method will never generate an estimate less than 1 (the mean is undefined for p < 1 for the Yule distribution).

    2. The method of zero frequency and sample mean:

        phat = (1/(1 - f1/N)/((1/(1-f1/N) - (1/xbar) - 1)

      with N and f1 denoting the sample size and the observed first frequency.

      This method is only useful for values of the shape parameter > 1.

    3. The method of maximum likelihood:

      The maximum likelihood estimate is the solution of the following equation:

        N/(C*(C-1)) - SUM[k=2 to lambda][V(k)/(C+k-1)] = 0

      with N denoting the sample size, Vk the cumulative frequency from k upwards, and lamda the maximum frequency.

      Then the parameter, p, of the Yule distribution is

        p = C - 1.

      These estimation methods are discussed in detail in the Irwin article (see the Reference section belows).

Default:
    None
Synonyms:
    None
Related Commands:
    YULCDF = Compute the Yule cumulative distribution function.
    YULPPF = Compute the Yule percent point function.
    WARPDF = Compute the Waring probability density function.
    BBNPDF = Compute the beta-binomial probability density function.
    GEOPDF = Compute the geometric probability density function.
    NBPDF = Compute the negative binomial probability density function.
    HYPPDF = Compute the hypergeometric probability density function.
Reference:
    "Discrete Univariate Distributions", Second Edition, Johnson, Kotz, and Kemp, John Wiley & Sons, 1994 (pp. 274-279).

    "Mathematcs in Medical and Biological Statistics", J. O. Irwin, Journal of the Royal Statistical Society, A, 1963, pp. 1-44.

Applications:
    Distributional Modeling
Implementation Date:
    2004/4
Program:
     
    Y1LABEL Probability
    X1LABEL X
    LABEL CASE ASIS
    X1LABEL DISPLACEMENT 12
    Y1LABEL DISPLACEMENT 12
    MULTIPLOT 2 2
    MULTIPLOT CORNER COORDINATES 0 0 100 100
    XTIC OFFSET 0.5 0.5
    LINE BLANK
    SPIKE ON
    TITLE AUTOMATIC
    X1LABEL X
    Y1LABEL PROBABILITY
    TITLE SIZE 3
    PLOT YULPDF(X,0.5) FOR X = 0 1 50
    PLOT YULPDF(X,1) FOR X = 0 1 50
    PLOT YULPDF(X,1.5) FOR X = 0 1 50
    PLOT YULPDF(X,2) FOR X = 0 1 50
    END OF MULTIPLOT
        
    plot generated by sample program

Date created: 7/7/2004
Last updated: 7/7/2004
Please email comments on this WWW page to alan.heckert@nist.gov.