Hi,
I am getting a warning message when I am fitting a generalized mixed model
(mod_2) and I don't understand why because when I add just an interaction
factor the model works perfectly (mod_1).
Does anyone know what it happpens ?
Thanks,
Aïda
>
mod_1<-lmer(sur15~soeviv15_4plus+frviv15_4plus+frat_15death+dad_class_new+soeviv15_4plus:dad_class_new
+frviv15_4plus:dad_class_new+
aine+ageatmumdeath+(1|newparish)+(1|mumID)+(1|cohorte),
+ weights=prop_off_tracked_to_age_15,family=poisson,
data=basepreindust2_sur15,RMEL=F)
> mod_1
Generalized linear mixed model fit by the Laplace approximation
Formula: sur15 ~ soeviv15_4plus + frviv15_4plus + frat_15death + dad_class_new
+ soeviv15_4plus:dad_class_new + frviv15_4plus:dad_class_new + aine +
ageatmumdeath + (1 | newparish) + (1 | mumID) + (1 |cohorte)
Data: basepreindust2_sur15
AIC BIC logLik deviance
2333 2416 -1151 2303
Random effects:
Groups Name Variance Std.Dev.
mumID (Intercept) 0.0470305 0.216865
newparish (Intercept) 0.0027002 0.051964
cohorte (Intercept) 0.0011787 0.034332
Number of obs: 1924, groups: mumID, 1293; newparish, 5; cohorte, 3
Fixed effects:
Estimate Std. Error z value Pr(>|z|)
(Intercept) 1.3013435 0.0570543 22.809 < 2e-16 ***
soeviv15_4plus -0.0608455 0.0191099 -3.184 0.00145 **
frviv15_4plus -0.0075483 0.0191727 -0.394 0.69380
frat_15death 0.0007571 0.0073176 0.103 0.91760
dad_class_new3 -0.0983394 0.0404488 -2.431 0.01505 *
dad_class_new4 -0.1401969 0.0744017 -1.884 0.05952 .
aine1 0.0139884 0.0389805 0.359 0.71970
ageatmumdeath 0.0019492 0.0009732 2.003 0.04519 *
soeviv15_4plus:dad_class_new3 0.0839922 0.0277758 3.024 0.00250 **
soeviv15_4plus:dad_class_new4 0.1014875 0.0560405 1.811 0.07015 .
frviv15_4plus:dad_class_new3 0.0020857 0.0287243 0.073 0.94212
frviv15_4plus:dad_class_new4 -0.0211242 0.0616165 -0.343 0.73172
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Correlation of Fixed Effects:
(Intr) sv15_4 fr15_4 frt_15 dd_c_3 dd_c_4 aine1 agtmmd s15_4:__3
s15_4:__4 f15_4:__3
sovv15_4pls
-0.237
frvv15_4pls -0.253
-0.139
frat_15deth -0.237 -0.154
-0.218
dd_clss_nw3 -0.303 0.214 0.270
0.072
dd_clss_nw4 -0.228 0.117 0.145 0.072
0.247
aine1 -0.155 0.247 0.264 -0.153 -0.039
-0.028
ageatmumdth -0.511 0.128 0.123 -0.273 -0.028 0.049
-0.049
svv15_4:__3 0.091 -0.608 0.182 -0.044 -0.402 -0.084 0.041
0.043
svv15_4:__4 0.061 -0.299 0.095 -0.029 -0.086 -0.481 0.039 0.001
0.214
frv15_4:__3 0.129 0.158 -0.592 0.024 -0.460 -0.107 0.022 -0.009 -0.219
-0.052
frv15_4:__4 0.058 0.077 -0.274 0.015 -0.088 -0.470 0.023 -0.006 -0.050
-0.047 0.186
>
mod_2<-lmer(sur15~soeviv15_4plus+frviv15_4plus+frat_15death+dad_class_new
+frviv15_4plus:dad_class_new+
aine+ageatmumdeath+(1|newparish)+(1|mumID)+(1|cohorte),
weights=prop_off_tracked_to_age_15,family=poisson,
data=basepreindust2_sur15,RMEL=F,control = list(msVerbose = 1))
0: 3767.6598: 1.33869 0.0832467 0.0644826 1.30576 -0.0220911 -0.0188889
0.00561227 -0.0367862 -0.0735001 0.00785921 0.00209046 0.0218301 -0.0103924
1: 3757.7897: 1.33703 0.0832908 0.0645599 1.30563 -0.0222456 -0.0188829
0.00518377 -0.0368487 -0.0735110 0.00787140 -0.00247635 0.0218423 -0.0103827
2: 3745.9984: 1.32280 0.0828516 0.0641048 1.30518 -0.0245513 -0.0190723
0.00486308 -0.0370936 -0.0735605 0.00842523 0.00231376 0.0219140 -0.0103036
3: 3234.0042: 0.977325 0.0741097 0.0555502 1.29274 -0.0734156 -0.0214680
-0.0100042 -0.0436855 -0.0748839 0.0199044 -0.00206090 0.0242118 -0.00837900
4: 3165.8228: 0.927184 0.0736596 0.0552550 1.40622 -0.101104 0.0337682
-0.0185303 -0.0464208 -0.0952062 0.0185455 -0.00162160 0.0237606 -0.00121856
5: 3143.7562: 0.903731 0.0729734 0.0544866 1.40540 -0.0967740 0.0325119
-0.0165578 -0.0468900 -0.0953327 0.0183130 -0.00648733 0.0233998 -0.00130747
6: 3115.6489: 0.898189 0.0727692 0.0546796 1.38913 -0.0885700 0.0187058
-0.0139194 -0.0467760 -0.0928831 0.0186559 -0.00134163 0.0210460 -0.00262605
7: 3082.2905: 0.872730 0.0719408 0.0555959 1.36488 -0.0710244 -0.00898995
-0.00997794 -0.0467335 -0.0899782 0.0190971 -0.00513307 0.0144395 -0.00494046
8: 2895.4929: 0.679086 0.0661372 0.0891088 1.91069 -0.146248 0.0705396
-0.0664482 -0.0517994 -0.202402 0.0177356 -0.00630075 -0.0654173 0.0199159
9: nan: -0.174440 0.823699 0.851474 -3.66912 0.0283103 -0.118455
0.193071 -1.35997 0.103926 -0.0112525 -0.262753 -0.414740 -27.2429
10: nan: -0.174440 0.823699 0.851474 -3.66912 0.0283103 -0.118455
0.193071 -1.35997 0.103926 -0.0112525 -0.262753 -0.414740 -27.2429
Message d'avis :
In mer_finalize(ans) :
Cholmod warning 'not positive definite' at
file:../Cholesky/t_cholmod_rowfac.c, line 432
> sessionInfo()
R version 2.11.1 (2010-05-31)
i386-pc-mingw32
locale:
[1] LC_COLLATE=French_France.1252 LC_CTYPE=French_France.1252
LC_MONETARY=French_France.1252 LC_NUMERIC=C
[5] LC_TIME=French_France.1252
attached base packages:
[1] grDevices datasets splines graphics stats tcltk utils
methods base
other attached packages:
[1] Design_2.3-0 MASS_7.3-6 languageR_1.0 zipfR_0.6-5
coda_0.13-5 nlme_3.1-96 lme4_0.999375-33 Matrix_0.999375-39
[9] lattice_0.18-8 svSocket_0.9-48 TinnR_1.0.3 R2HTML_2.1
Hmisc_3.8-1 survival_2.35-8
loaded via a namespace (and not attached):
[1] cluster_1.12.3 grid_2.11.1 svMisc_0.9-57 tools_2.11.1
Une messagerie gratuite, garantie à vie et des services en plus, ça vous tente ?
Je crée ma boîte mail www.laposte.net
[[alternative HTML version deleted]]
