Turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

- Home
- /
- Analytics
- /
- Stat Procs
- /
- Re: Likelihood Ratio Test in Mixed Model - Proc Mixed

Options

- RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Mute
- Printer Friendly Page

🔒 This topic is **solved** and **locked**.
Need further help from the community? Please
sign in and ask a **new** question.

- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content

Posted 03-31-2018 07:27 AM
(6190 views)

The likelihood ratio test of comparing reduced model with full model differs by fixed factor result to chi-square distribution of zero degree of freedom.

/* reduced model */

proc mixed method = ml;

class block gen;

model rtwt = /ddfm = kr;

random block gen;

run;

/* full model*/

proc mixed method = ml;

class block gen;

model rtwt = prop_hav/ddfm = kr;

random block gen;

run;

There are 3 degree of freedom from reduced model - block variance, genotype variance, and residual variance. The same degree of freedom for full model despite inclusion of prop_hav as covariate. The difference in their -2 loglikelihood has zero degree of freedom under chi-distribution. Please could anyone guide me on how to compare these model to ascertain if the full model is significantly different from reduced model.

1 ACCEPTED SOLUTION

Accepted Solutions

- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content

You misunderstand the method. Yes, the two models have the same number of parameters associated with random effects (variances or covariances), but they have different numbers of parameters associated with fixed effects.

You seek to assess the significance of a continuous-scale fixed effects factor (prop_hav) using a likelihood ratio test. The full model has 5 parameters: intercept, *slope for prop_hav,* and 3 variances (rep, gen, residual). The reduced model has 4 parameters: intercept, and 3 variances (rep, gen, residual). The difference in the number of parameters between the two models is 5 - 4 = 1.

6 REPLIES 6

- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content

- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content

Thanks for the reply. Please have a look at the SAS script I posted. Both reduced and full model have the same number of covariance parameters which is 3.

/* reduced model*/;

proc mixed method=ml;

class rep gen;

model rtwt=/ddfm=kr;

random rep gen;

run;

quit;

/* Full model */;

proc mixed method=ml;

class rep gen;

model rtwt= prop_hav /ddfm=kr;

random rep gen;

run;

quit;

As you can see, the full model has prop_hav as a covariate. The only difference between the model is the fixed factor which is covariate. The number of parameters in each case is 3: rep variance, gen variance, and residual variance. This means the difference of -2 log likelihood with have chi-square distribution with zero degree of freedom (3-3) and NOT 1 degree of freedom.

Thanks

- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content

You misunderstand the method. Yes, the two models have the same number of parameters associated with random effects (variances or covariances), but they have different numbers of parameters associated with fixed effects.

You seek to assess the significance of a continuous-scale fixed effects factor (prop_hav) using a likelihood ratio test. The full model has 5 parameters: intercept, *slope for prop_hav,* and 3 variances (rep, gen, residual). The reduced model has 4 parameters: intercept, and 3 variances (rep, gen, residual). The difference in the number of parameters between the two models is 5 - 4 = 1.

- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content

Thanks a lot for your informative response. Your response helps a lot to figure out the solution.

- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content

I'm pleased you found it helpful.

- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content

Hi,

Just posted questions on externally conditional studentized residual from proc mixed. I would appreciate your guide and assistance on this.

Thanks

**Available on demand!**

Missed SAS Innovate Las Vegas? Watch all the action for free! View the keynotes, general sessions and 22 breakouts on demand.

What is ANOVA?

ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.

Find more tutorials on the SAS Users YouTube channel.