Distribution plot from proc univariate

Miracle — Thu, 19 May 2016 06:44:06 GMT

Dear Sir or Madam,

Can I please ask what do this errors mean? How can I resolve it? Thank you very much.

560  proc univariate data=temp;
561      var occ;
562      histogram / midpoints=0 to 600 by 50
563      lognormal
564     weibull
565      gamma;
566      inset n mean(5.3) std='Std Dev'(5.3) skewness(5.3)
567      / pos = ne  header = 'Summary Statistics';
578  run;

NOTE: Since a threshold parameter (THETA) was not specified for the lognormal fit for occ, a
      zero threshold is assumed.
ERROR: The smallest value of occ is less than or equal to the threshold parameter (THETA) for
       the lognormal fit.
NOTE: The lognormal threshold parameter (THETA) must be less than 0.
NOTE: The lognormal curve was not drawn because estimates could not be computed for all
      parameters.
NOTE: Since a threshold parameter (THETA) was not specified for the Weibull fit for occ, a
      zero threshold is assumed.
ERROR: The smallest value of occ is less than or equal to the threshold parameter (THETA) for
       the Weibull fit.
NOTE: The Weibull threshold parameter (THETA) must be less than 0.
NOTE: The Weibull curve was not drawn because estimates could not be computed for all
      parameters.
NOTE: Since a threshold parameter (THETA) was not specified for the gamma fit for occ, a zero
      threshold is assumed.
ERROR: The smallest value of occ is less than or equal to the threshold parameter (THETA) for
       the gamma fit.
NOTE: The gamma threshold parameter (THETA) must be less than 0.
NOTE: The gamma curve was not drawn because estimates could not be computed for all parameters.
NOTE: At least one W.D format was too small for the number to be printed. The decimal may be
      shifted by the "BEST" format.
NOTE: PROCEDURE UNIVARIATE used (Total process time):
      real time           0.19 seconds
      cpu time            0.04 seconds

Re: Distribution plot from proc univariate

Rick_SAS — Thu, 19 May 2016 09:49:28 GMT

In three=parameter families that have a threshold parameters, the data must be strictly larger than the values of the threshold parameter. The reason is that the distributions you are fitting have a term that looks like

log( x[i] - theta )

(x[i] - theta)**c.

The LOG function is undefined when its argument is zero or negative. Similarly, the "raise to power function" is undefined for negative arguments when c is not an integer. (For example, c=1/2 is the square roo function.)

The most likely reason for these errors is that you have data values that are zero. Your choices include:

Model the data by using different distributions
Specify a value for the threshold parameters that is strictly less than the smallest data value
Delete the bad data values if they are coding errors

topic Distribution plot from proc univariate in Graphics Programming

Distribution plot from proc univariate

Re: Distribution plot from proc univariate