My confidence bands look fine, so something is going on with my (or his!) estimation of that second autocorrelation. I thought this might be due to randomness, so I tried it several times, only to get the same (apparently wrong) result. On my AC graph, the second autocorrelation is outside of the confidence bands (only the first one should be significant, according to Becketti). This is my code, with the same sample size and seed the author uses: It's supposed to show the autocorrelation and partial autocorrelation functions of a simulated ARMA(1,1) process \ with \ and \. I have just reproduced figures 7.9 through 7.12 in Sean Becketti's book, but I can't seem to get 7.13 correctly. Mi predict myprediction using myestimates Mi estimate, saving(myestimates, replace): xtmixed weight week gender || id: week, covariance(unstructured) Mi estimate: xtmixed weight i.week i.gender || id: week, covariance(unstructured)
*/ So, I added gender (1= male,0 = female) to the fixed part of the random-effects coefficient model */ I would like to allow for a random slope on week, an unstructured covariance structure and that results are adjusted for gender. Mi impute monotone (regress) weight9 = cov1 cov2 cov3 weight8 gender, add(5) Replace weight9 = cond(runiform()<0.30.,weight9) */ generate 30% of missing data at random for week 9 */ generate 4 hypothetical auxiliary variables, 3 continuous and 1 binary (1 = male, 0 = female) 48 pigs were followed up for 9 weeks and their weight was recorded.
#STATA DEDLETE XLIST DOWNLOAD#
*/ Download the dataset used by Ruppert, Wand, and Carroll (2003) and Diggle et al. I will be extremely thankful for any help. If I have a variable, say "weight", measured over time (week=1,2,3,4,5,6,7,8 and 9), and need that predictions for variable weight at time 2 need to be adjusted for weight at time 1 (baseline), how do I do that ? Is there any straightforward, non-flamboyant way to obtain predictions with both fixed and random parts after -xtmixed- and multiple imputation?Ģ.
So, I've prepared a small script to make myself clearer. Rabe-Hesketh & Skrondal's book has been really useful, but it does not address multiple imputation. I've been trying to understand -xtmixed- and multiple imputation.
0 kommentar(er)
