7.00.02 - Results (statistics and p-values) - Aster Analytics

Teradata Aster® Analytics Foundation User GuideUpdate 2

Product
Aster Analytics
Release Number
7.00.02
Release Date
September 2017
Content Type
Programming Reference
User Guide
Publication ID
B700-1022-700K
Language
English (United States)
These results see the following R packages:

For a normal distribution, the function computes p-values from the modified statistics as implemented in the R package nortest.

For other continuous distributions:
  • KS

    The function computes p-values using approximation to the Kolmogorov-Smirnov distribution (assuming large sample size). Results are comparable to the ks.test() from the stats package in R.

  • CvM

    The function computes results as in the R package nortest.

  • AD

    The function computes results that are comparable to the ad.test() from the R package ADGofTest.

  • CHISQ

    Statistics are comparable to pearson.test() from the R package nortest, with a minor modification when data falls beyond the upper limit of distribution support. The function computes p-values using DOF = cell_size - 1.

For discrete distributions:
  • KS

    Results are comparable to the ks.test() from the R package dgof, when reference distribution is a 'stepfun' object.

  • CvM

    Statistics are comparable to the cvm.test() from the R package dgof, when reference distribution is a 'stepfun' object. P-values are not calculated.

  • AD

    Results are comparable to the ad.test() from the R package ADGofTest.

  • CHISQ

    Statistics are comparable to pearson.test() from the R package nortest, with a minor modification when data falls beyond the upper limit of distribution support. The function computes p-values using DOF = cell_size - 1.