On Feb 16, 2009, at 9:57 PM, Barker, Chris wrote:
>
> Hi, I have a list of fit objects (fit objects from HMISC functions)
"Hmisc", and they are actually probably from "Design"
functions.
>
>
> I create elements in the list in this way
>
> lrm.sumtot <- lrm( ae7bepn ~ trarm + sumtot , data=sd.fix)
>
> lrm.list[['lrm.sumtot']] <- lrm.sumtot
>
> And I can run (anova(lrm.sumtot))
>
> The following also gives the anova I'd expect
> zz <- lrm.list[['lrm.sumtot']];anova(zz)
>
>
> And similarly for the summary command
>
> I would prefer not to create a second list with "anova" objects
nor a
> third list with "summary" objects.
>
> Is there a way to use an sapply or lapply to prepare anova or summary
> for every element in my list of fit objects?
Note: these are are not nested models or even models built on the same
data:
> mod.list.objs <- list(cph.A.S.H.T,cph.A.S.H.T.G,cph.A.S.H.T.G.S.D)
> mapply(anova, mod.list.objs)
[[1]]
Wald Statistics Response: srv900
Factor Chi-Square
d.f. P
Age 17522.30
2 <.0001
Nonlinear 2608.09
1 <.0001
Sex 39.90
1 <.0001
BL_HDL.A (Factor+Higher Order Factors) 2485.88
6 <.0001
All Interactions 77.97
4 <.0001
Nonlinear (Factor+Higher Order Factors) 59.72
3 <.0001
BL_CHOLEST.A (Factor+Higher Order Factors) 1170.03
6 <.0001
All Interactions 77.97
4 <.0001
Nonlinear (Factor+Higher Order Factors) 226.03
3 <.0001
BL_HDL.A * BL_CHOLEST.A (Factor+Higher Order Factors) 77.97
4 <.0001
Nonlinear 75.20
3 <.0001
Nonlinear Interaction : f(A,B) vs. AB 75.20
3 <.0001
f(A,B) vs. Af(B) + Bg(A) 59.16
1 <.0001
Nonlinear Interaction in BL_HDL.A vs. Af(B) 59.16
2 <.0001
Nonlinear Interaction in BL_CHOLEST.A vs. Bg(A) 68.80
2 <.0001
TOTAL NONLINEAR 2968.98
6 <.0001
TOTAL NONLINEAR + INTERACTION 3515.27
7 <.0001
TOTAL 23438.25
11 <.0001
[[2]]
Wald Statistics Response: srv900
Factor Chi-Square
d.f. P
Age 10025.10
2 <.0001
Nonlinear 105.89
1 <.0001
Sex 158.10
1 <.0001
BL_HDL.A (Factor+Higher Order Factors) 144.11
6 <.0001
All Interactions 32.53
4 <.0001
Nonlinear (Factor+Higher Order Factors) 91.43
3 <.0001
BL_CHOLEST.A (Factor+Higher Order Factors) 267.75
6 <.0001
All Interactions 32.53
4 <.0001
Nonlinear (Factor+Higher Order Factors) 243.49
3 <.0001
BL_GGT.A 395.13
1 <.0001
BL_HDL.A * BL_CHOLEST.A (Factor+Higher Order Factors) 32.53
4 <.0001
Nonlinear 15.82
3 0.0012
Nonlinear Interaction : f(A,B) vs. AB 15.82
3 0.0012
f(A,B) vs. Af(B) + Bg(A) 1.46
1 0.2271
Nonlinear Interaction in BL_HDL.A vs. Af(B) 11.91
2 0.0026
Nonlinear Interaction in BL_CHOLEST.A vs. Bg(A) 2.01
2 0.3658
TOTAL NONLINEAR 474.30
6 <.0001
TOTAL NONLINEAR + INTERACTION 475.48
7 <.0001
TOTAL 11723.84
12 <.0001
[[3]]
Wald Statistics Response: srv900
Factor Chi-Square
d.f. P
Age 9168.80
2 <.0001
Nonlinear 138.98
1 <.0001
Sex 48.89
1 <.0001
BL_HDL.A (Factor+Higher Order Factors) 85.17
6 <.0001
All Interactions 28.80
4 <.0001
Nonlinear (Factor+Higher Order Factors) 71.08
3 <.0001
BL_CHOLEST.A (Factor+Higher Order Factors) 301.41
6 <.0001
All Interactions 28.80
4 <.0001
Nonlinear (Factor+Higher Order Factors) 243.65
3 <.0001
BL_GGT.A 701.68
2 <.0001
Nonlinear 256.32
1 <.0001
BL_TRIGLYC.A 0.33
1 0.5647
BP.B 1.27
1 0.2596
BP.A 546.29
1 <.0001
BL_HDL.A * BL_CHOLEST.A (Factor+Higher Order Factors) 28.80
4 <.0001
Nonlinear 11.82
3 0.0080
Nonlinear Interaction : f(A,B) vs. AB 11.82
3 0.0080
f(A,B) vs. Af(B) + Bg(A) 5.61
1 0.0178
Nonlinear Interaction in BL_HDL.A vs. Af(B) 10.35
2 0.0057
Nonlinear Interaction in BL_CHOLEST.A vs. Bg(A) 5.65
2 0.0594
TOTAL NONLINEAR 677.48
7 <.0001
TOTAL NONLINEAR + INTERACTION 679.59
8 <.0001
TOTAL 11874.22
16 <.0001
mapply(summary, ...) also works
--
David Winsemius>
>
>
> Thank you in advance
>
> Chris Barker, Ph.D.
> Statistical Consultant
> 650 384 8617
>
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> 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.