[R] max possible rsquare

[Uwe Ligges]

On 13.07.2011 14:42, Rilana Prenger wrote:
> Dear all,
>
> I have a question regarding the output of the coxph function. What does the
> 'max possible' exactly mean in the output below? Many thanks.
>
>                 coef exp(coef) se(coef) robust se      z Pr(>|z|)
> smocc_zyban      -0.4384    0.6451   0.8667    0.9473 -0.463    0.644
> self              1.1857    3.2728   0.1405    0.1443  8.216 2.22e-16 ***
> smocc_zyban:self  0.1423    1.1529   0.2201    0.2382  0.598    0.550
> ---
> Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
>
>                   exp(coef) exp(-coef) lower .95 upper .95
> smocc_zyban         0.6451     1.5502    0.1008     4.130
> self                3.2728     0.3055    2.4666     4.343
> smocc_zyban:self    1.1529     0.8673    0.7229     1.839
>
> Rsquare= 0.116   (max possible= 0.821 )

This is what can be achieved maximally with the most extreme possible 
value of the test statistics which is used to calculate the Rsquare. 
That means, 0.821 (rather than 1) is the "perfect fit" given the likelihood.

Uwe Ligges



> Likelihood ratio test= 137  on 3 df,   p=0
> Wald test            = 114.3  on 3 df,   p=0
> Score (logrank) test = 141.3  on 3 df,   p=0,   Robust = 70.98  p=2.665e-15
>
> --
> View this message in context: http://r.789695.n4.nabble.com/max-possible-rsquare-tp3664925p3664925.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.