Getting the f statistic from an anova output of aster models

[M Abdur-Razak]

Is there a way to do this? The names in an anova output are "Model Df" "Model Dev" "Df" "Deviance" and "P(>|Chi|)". Is there a way to extra the f statistic from this?

Thanks

[M Abdur-Razak]

extract**

[geyer]

There is no F statistic anywhere in aster theory (nor in GLM theory for that matter).  F distributions for test statistics require the assumption of exact normality and homoscedasticity of errors -- the "usual" assumptions for linear models (including ANOVA).  The R generic function anova is use to do not only ANOVA for linear models but also "analysis of deviance" for arbitrary models (that is, likelihood ratio tests).  That is what aster is doing.  The "deviance" (= likelihood ratio test statistic) is reported by the anova function.

[M Abdur-Razak]

I understand. Thanks for taking the time.