Pearson’s Chi-squared Statistic - Teradata Vantage

Machine Learning Engine Analytic Function Reference

Product
Teradata Vantage
Release Number
9.02
9.01
2.0
1.3
Published
February 2022
Language
English (United States)
Last Update
2022-02-10
dita:mapPath
rnn1580259159235.ditamap
dita:ditavalPath
ybt1582220416951.ditaval
dita:id
B700-4003
lifecycle
previous
Product Category
Teradata Vantage™

The deviance generalizes the sum of squared errors. Another generalization of sum of squared errors is Pearson’s chi-squared statistic. Given a generalized linear model with responses yi, weights wi, fitted means μi, variance function v(μ) and dispersion φ = 1, the Pearson goodness-of-fit statistic is



If the fitted model is correct and the observations yi are approximately normal, X2 is approximately distributed as X2on the residual degrees of freedom for the model. Both the deviance and the generalized Pearson X2 have exact X2 distributions for Normal-theory linear models (assuming of course that the model is true), and asymptotic results are available for the other distributions. The deviance has a general advantage as a measure of discrepancy in that it is additive for nested sets of models if maximum-likelihood estimates are used, whereas X2 in general is not. However, X2 may sometimes be preferred because of its more direct interpretation.

The GLM function computes the Pearson’s goodness of fit.