<p>请教哪位高人，</p><p>Greene书中 或者 Stata帮助中给出的关于Nested logit model的例题，都是使用的alternative-specific and alternative-invariant混合的data，stata中nlogit命令也是针对着这种数据格式的。哪位大侠做过只使用alternative-invariant 数据做的nested分析呢?</p><p>我觉得实质应该是相同的，但是我在计算中总是报错。</p><p>FIML估计时，stata的错误报告是<span lang="EN-US" style="FONT-SIZE: 16pt;"><font face="Times New Roman" size="3">variable&nbsp;** has replicate levels for one or more cases; this is not allowed</font></span></p><p><span lang="EN-US" style="FONT-SIZE: 16pt;"><font face="Times New Roman"><font size="3">LIML估计时，又报告</font><font size="2"><span lang="EN-US" style="FONT-SIZE: 16pt; FONT-FAMILY: &quot;Times New Roman&quot;; mso-fareast-font-family: 宋体; mso-font-kerning: 1.0pt; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA;">IV predicts success perfectly, so dropped。因为IV （inclusive value）是我计算的，所以我怀疑很可能是我的错，但又不知道错在哪里</span></font></font></span></p><p><span lang="EN-US" style="FONT-SIZE: 16pt;"><font face="Times New Roman" size="2"><span lang="EN-US" style="FONT-SIZE: 16pt; FONT-FAMILY: &quot;Times New Roman&quot;; mso-fareast-font-family: 宋体; mso-font-kerning: 1.0pt; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA;">大家多多指教</span></font></span></p>