3
$\begingroup$

I want to use linear mixed effect model for a set of data.

After using lme4 package and lmer() function and fitting model, I want to validate my model for other period time. But I don't know how to validate my model and generate predicted values.

Do you have any ideas about useful validation method? Is there any command about validation in R software?


Thanks for your useful comments.

I fitted model to one set of data and now I want to check its performance for an other data set. I want to plot figure by "predicted value" and "observed value" and calculate R^2 to evaluate model performance.

Thanks

$\endgroup$
  • 3
    $\begingroup$ What do you mean by "validate"? One meaning could be "check assumptions"; another "check that it fits other data". Possible more meanings that that, too. $\endgroup$ – Peter Flom Nov 9 '13 at 20:54
  • 2
    $\begingroup$ agree with @PeterFlom that this question is a bit too vague. In the meanwhile, you could look at ?predict.merMod ... $\endgroup$ – Ben Bolker Nov 9 '13 at 21:05
  • $\begingroup$ Thanks for your useful comments. I fitted model to one set of data and now I want to check its performance for an other data set. I want to plot figure by "predicted value" and "observed value" and calculate R^2 to evaluate model performance. Thanks $\endgroup$ – saeed Nov 10 '13 at 9:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.