<p>easy to learn:</p><p><a href="http://www.ats.ucla.edu/stat/sas/faq/simple_nlmixed.htm"><!-- m --><a class="postlink" href="http://www.ats.ucla.edu/stat/sas/faq/simple_nlmixed.htm">http://www.ats.ucla.edu/stat/sas/faq/simple_nlmixed.htm</a><!-- m --></a></p><p>[ transforming the multinomial model into a multivariate binary model.] - 'problem disturb me for long time'</p><p><<a note on the estimation of the multinomial logistic model with correlated responses in sas >> page 9 </p><p> </p> <a href="http://hi.baidu.com/dwdus/blog/item/29383b1320beb51d5aaf5385.html">阅读全文</a>