Hi everyone, I have carried out a multiple imputation in R using Amelia II and have created 5 multiply imputed datasets. The purpose of my research is to fit a Poisson Model to the data to estimate numbers of hospital admissions.
Now that I have 5 completed datasets and I have to pool all the 5 datasets to get one combined output for a poisson model. I have checked previous queries about combining MI datasets and they are quite dated and none really related to my problem. I have been tirelessly working with zelig and unfortunately i am getting real stuck! here is my R code: ###these are my 5 imputed datasets#### d.1 <- read.table("imp1.csv", header=TRUE,sep=",") d.2 <- read.table("imp2.csv", header=TRUE,sep=",") d.3 <- read.table("imp3.csv", header=TRUE,sep=",") d.4 <- read.table("imp4.csv", header=TRUE,sep=",") d.5 <- read.table("imp5.csv", header=TRUE,sep=",") #####my zelig code##### model3=zelig(DKA.counts~age+as.factor(sex)+bmi_value+hba1c_value,model="poisson",data=mi(d.1,d.2,d.3,d.4,d.5)) summary(model3) Till so far it is fine... but after this I get separate output for each dataset! I have tried and tried to find the problem but i still cannot get the combined output. Can anyone please tell me what i'm doing wrong!?! #####output#### $d.1 Call: glm(formula = formula, family = poisson(), data = data, weights = weights, model = F) Deviance Residuals: Min 1Q Median 3Q Max -1.9410 -0.6990 -0.4136 0.1423 10.3703 Coefficients: Estimate Std. Error z value Pr(>|z|) (Intercept) -0.638545 0.097690 -6.536 6.3e-11 *** age 0.033006 0.004625 7.136 9.6e-13 *** as.factor(sex)1.5 0.672653 0.289634 2.322 0.0202 * as.factor(sex)2 0.131818 0.027845 4.734 2.2e-06 *** bmi_value -0.007784 0.003794 -2.051 0.0402 * hba1c_value 0.087617 0.006812 12.862 < 2e-16 *** --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 (Dispersion parameter for poisson family taken to be 1) Null deviance: 3656.9 on 2943 degrees of freedom Residual deviance: 3299.6 on 2938 degrees of freedom AIC: 10307 Number of Fisher Scoring iterations: 5 $d.2 : : $d.3 : : $d.4 : : $d.5 Call: glm(formula = formula, family = poisson(), data = data, weights = weights, model = F) Deviance Residuals: Min 1Q Median 3Q Max -1.8925 -0.6973 -0.4194 0.1321 10.3741 Coefficients: Estimate Std. Error z value Pr(>|z|) (Intercept) -0.696957 0.097539 -7.145 8.97e-13 *** age 0.029981 0.004636 6.467 1.00e-10 *** as.factor(sex)1.5 0.671931 0.289643 2.320 0.0203 * as.factor(sex)2 0.122598 0.027794 4.411 1.03e-05 *** bmi_value -0.002243 0.003719 -0.603 0.5465 hba1c_value 0.086084 0.006774 12.709 < 2e-16 *** --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 (Dispersion parameter for poisson family taken to be 1) Null deviance: 3656.9 on 2943 degrees of freedom Residual deviance: 3308.9 on 2938 degrees of freedom AIC: 10316 Number of Fisher Scoring iterations: 5 Many thanks in advance!! -- View this message in context: http://r.789695.n4.nabble.com/combined-output-with-zelig-is-not-working-tp4647566.html Sent from the R help mailing list archive at Nabble.com. ______________________________________________ R-help@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.