Run Time

Welcome to the forum for REALCOM users. Feel free to post your question about REALCOM here. The Centre for Multilevel Modelling take no responsibility for the accuracy of these posts, we are unable to monitor them closely. Do go ahead and post your question and thank you in advance if you find the time to post any answers!

Go REALCOM (Developing multilevel models for REAListically COMplex social science data) >> http://www.bristol.ac.uk/cmm/software/realcom/
Post Reply
NilsSwindell
Posts: 1
Joined: Tue Jul 25, 2017 1:43 pm

Run Time

Post by NilsSwindell »

Hi, I am using Reacom impute for the first time. I have got a data set with 17,000 cases and I have 14 variables in my model of interest.
In my imputation model, I have specified a burn in of 1500 iterations and 500 iteration between each of the 10 imputations. As a result, the model has taken over 7 days to run and is only half way though. Has anyone got any advise on how long such a model should take and if anything can be done to speed it up?

Many thanks,

Nils
ChrisCharlton
Posts: 1348
Joined: Mon Oct 19, 2009 10:34 am

Re: Run Time

Post by ChrisCharlton »

The time that it takes to run the model will depend on a number of factors including the response types in the imputation model, so it is difficult to provide a typical run time. Unfortunately there is not much that can be done to speed up running the model with Realcom, however if you have access to Stat-JR, either as an MLwiN user or as a UK based academic, then you might want to try the imputation templates that we have implemented for it (see http://www.bristol.ac.uk/cmm/research/m ... jr_missing). These cover a lot of the same model types as Realcom and generally are able to fit them several orders of magnitude faster.
RitaPickett
Posts: 5
Joined: Sat Nov 18, 2017 5:35 am

Re: Run Time

Post by RitaPickett »

One of the main factor in which speed of a system is considered is response time. In technology, response time is the time a system or functional unit takes to react to a given input. Imputation theory is constantly developing and thus requires consistent attention to new information regarding the subject. There have been many theories embraced by scientists to account for missing data but the majority of them introduce large amounts of bias. You can use the template provided by Chris. If you are MLwin user then it is more easy as MLwiN is a statistical software package for fitting multilevel models. It uses both maximum likelihood estimation and Markov Chain Monte Carlo methods. MLwiN is based on an earlier package, but with a graphical user interface.
brucedeniger
Posts: 4
Joined: Wed Dec 06, 2017 6:12 am

Re: Run Time

Post by brucedeniger »

A Stata program is available for import and export to the Realcom Impute software.

The Stata program can be installed by typing the following (one line) in Stata’s command prompt:

net install realcomImpute, from (https://raw.githubusercontent.com/jwb13 ... te/master/) replace
Post Reply