[R] doubt on survreg.
Ronaldo Reis Jr.
chrysopa at insecta.ufv.br
Mon Dec 23 21:00:10 CET 2002
Hi all,
I make a censored model using survreg from survival package. But the deviance
values are too high, it is possible that must be a error in my model?
> summary(completo)
conc peso dv w temp
Min. :0.7 Min. : 5.00 Min. : 1.00 Min. :0.0000 Min. :22.70
1st Qu.:2.3 1st Qu.: 9.00 1st Qu.: 4.00 1st Qu.:0.0000 1st Qu.:24.20
Median :3.9 Median :11.00 Median :11.00 Median :0.0000 Median :25.00
Mean :3.9 Mean :11.21 Mean :12.90 Mean :0.2727 Mean :25.22
3rd Qu.:5.5 3rd Qu.:13.00 3rd Qu.:22.25 3rd Qu.:1.0000 3rd Qu.:26.30
Max. :7.1 Max. :23.00 Max. :29.00 Max. :1.0000 Max. :30.00
sp
Erato :66
Ethila:66
> m0 <- survreg(Surv(dv,w)~1)
> m0
Call:
survreg(formula = Surv(dv, w) ~ 1)
Coefficients:
(Intercept)
3.268812
Scale= 0.08118577
Loglik(model)= -86.5 Loglik(intercept only)= -86.5
n= 132
> m1 <- update(m0,.~. +sp+sp:conc+conc+peso+temp)
> m1
Call:
survreg(formula = Surv(dv, w) ~ sp + conc + peso + temp + sp:conc)
Coefficients:
(Intercept) spEthila conc peso temp
6.16976497 0.47406628 -0.17860548 0.08421499 -0.16791187
spEthila:conc
-0.07836046
Scale= 0.0809617
Loglik(model)= -1.088904e+40 Loglik(intercept only)= -86.5
Chisq= -2.177807e+40 on 5 degrees of freedom, p= 1
n= 132
>
Thanks for all and happy new year.
Ronaldo
--
In marriage, as in war, it is permitted to take every advantage of the enemy.
--
| //|\\ [*****************************][*******************]
|| ( õ õ ) [Ronaldo Reis Júnior ][PentiumIII-600 ]
| V [ESALQ/USP-Entomologia, CP-09 ][HD: 30 + 10 Gb ]
|| / l \ [13418-900 Piracicaba - SP ][RAM: 128 Mb ]
| /(lin)\ [Fone: 19-429-4199 r.229 ][Video: SiS620-8Mb ]
||/(linux)\ [chrysopa at insecta.ufv.br ][Modem: Pctel-onboar]
|/ (linux) \[ICQ#: 5692561 ][Kernel: 2.4.18 ]
|| ( x ) [*****************************][*******************]
||| _/ \_Powered by Gnu/Debian Woody D+:) | Lxuser#: 205366
More information about the R-help
mailing list