Search found 227 matches

by mneely
Wed May 22, 2019 9:11 am
Forum: How Do I...?
Topic: Missing covariate values
Replies: 4
Views: 19

Re: Missing covariate values

PMstep() will do a linear regression backwards elimination (could do forward/backward) of all covariates vs. all parameters and report the P values for regression coefficients based on retention in the linear model. It uses AIC to discard or retain covariates. PMstep is based on the R "step" functio...
by mneely
Fri May 17, 2019 10:35 am
Forum: How Do I...?
Topic: Missing covariate values
Replies: 4
Views: 19

Re: Missing covariate values

If you have multiple interacting drugs possible and you want to model all of them, then set a covariate to 0 or 1 for each drug and use an exponential model. #PRI Ke0,0,5 mod1,-1,1 mod2,-1,1 mod3,-1,1 mod4,-1,1 ... #COV INT1 INT2 INT3 INT4 ... #SEC Ke = Ke0 * exp(mod1*INT1 + mod2*INT2 + mod3*INT3 + ...
by mneely
Fri May 17, 2019 10:29 am
Forum: How Do I...?
Topic: Missing covariate values
Replies: 4
Views: 19

Re: Missing covariate values

You could do something like INT = 0 or 1 as a covariate for receipt of an interacting drug and then have your model do this: #PRI Ke1,0,5 Ke2,0,5 ... #COV INT #SEC Ke = Ke1*(1-INT) + Ke2*INT This will set Ke = Ke1 when INT=0 (no interacting drug) and Ke = Ke2 when INT=1 (interacting drug present). D...
by mneely
Mon May 06, 2019 2:15 pm
Forum: How Do I...?
Topic: Error when include many variables
Replies: 3
Views: 50

Re: Error when include many variables

My first question was whether you had been able to run a simpler model, even one from the examples available on our website. Just want to make sure you have a working installation.
by mneely
Mon May 06, 2019 2:13 pm
Forum: How Do I...?
Topic: Three comp model
Replies: 1
Views: 34

Re: Three comp model

I don't see an error with your output block. Can you run it anyway without PMcheck? NPrun(...,nocheck=T) will suppress PMcheck.
by mneely
Mon May 06, 2019 2:10 pm
Forum: How Do I...?
Topic: Error npagdriv.f - no such file
Replies: 4
Views: 46

Re: Error npagdriv.f - no such file

Hi David, Something isn't allowing the run to start, which is why there isn't an npagdriv.f file. If you look at the DOS/Terminal window before closing it, there might be an error message embedded there. I don't see any errors in your #DIFF block, so it is likely elsewhere. Do you have definitions f...
by mneely
Wed Apr 17, 2019 10:23 am
Forum: How Do I...?
Topic: Error when include many variables
Replies: 3
Views: 50

Re: Error when include many variables

Two things: 1) Did it ever run as a simpler model? 2) Make sure you define the purpose for your model. If you want to simulate, say for a PTA, and you truly believe that all those covariates are related to your model parameters, then it makes sense to include them all. However, you will likely be fi...
by mneely
Fri Apr 05, 2019 9:42 am
Forum: How Do I...?
Topic: Compiler
Replies: 1
Views: 55

Re: Compiler

You should be able to type "1" and press <Return>.
by mneely
Fri Apr 05, 2019 9:41 am
Forum: How Do I...?
Topic: CI 95% of the PTA simulations
Replies: 1
Views: 45

Re: CI 95% of the PTA simulations

That's a good suggestion. We can add it. In the meantime, a quick google search for the 95% CI of a proportion reveals the formula p ± 1.96 * sqrt (p * (1-p) / n), where p is the PTA and n=1000 as there are always 1000 simulations per regimen in a PTA. You can get p from the summary() of the PMpta o...
by mneely
Fri Apr 05, 2019 9:32 am
Forum: How Do I...?
Topic: Differences in the VPC plots
Replies: 1
Views: 36

Re: Differences in the VPC plots

One VPC is in the npde package. I did not write the code for that. It is a standard VPC as described in the literature. I believe observations are time clustered using the mclust package for the purposes of calculating observation percentiles. The VPC made by plotting a PMsim object with real observ...