I&#39;m attempting to use AmeliaII in the following way: <br><br>vars_to_impute &lt;- gundata[,c(&quot;progun&quot;, &quot;egalitarianism&quot;, &quot;individualism&quot;,<br>&nbsp;&nbsp;&nbsp; &quot;crfear&quot;, &quot;victim&quot;, &quot;female&quot;, &quot;RACE&quot;, <br>

&nbsp;&nbsp;&nbsp; &quot;income&quot;, &quot;URBANKID&quot;, &quot;URBANNOW&quot;, <br>&nbsp;&nbsp;&nbsp; &quot;RELIGION&quot;, &quot;iss&quot;, &quot;democrat&quot;, &quot;conservative&quot;)]<br><br>imputed &lt;- amelia(data=vars_to_impute, p2s=2, noms=c(&quot;RACE&quot;,&quot;RELIGION&quot;,<div id="1fh5" class="ArwC7c ckChnd">
&quot;URBANKID&quot;,&quot;URBANNOW&quot;), outname=&quot;imputation&quot;, ords=c(&quot;democrat&quot;, &quot;conservative&quot;)<br>
<br>The problem that I run into is that in about half of the imputation
attempts fail due to non-invertible covariance matrices. I&#39;m curious if
there is any way to deal with this aside from removing the most highly
covariant variables?&nbsp; For example, given that it produces imputed data
about 1/5 of the time, would it be acceptable to use successful
imputations?&nbsp; E.g., can I just set m=100 and use as many imputations as
I like from the resulting set of ~25 imputations? <br>
<br>If I do need to remove the covariate variables, do you know of a simple way to check for that among a given set of variables?&nbsp; <br clear="all"><font color="#888888"><br></font></div>