topic Re: Finding the correct SD in PROC NLMIXED in SAS Programming https://communities.sas.com/t5/SAS-Programming/Finding-the-correct-SD-in-PROC-NLMIXED/m-p/651586#M195510 No one have any suggestion or idea for me????? Thu, 28 May 2020 23:20:24 GMT MahdiA110 2020-05-28T23:20:24Z Finding the correct SD in PROC NLMIXED https://communities.sas.com/t5/SAS-Programming/Finding-the-correct-SD-in-PROC-NLMIXED/m-p/649426#M194698 <P>Hello</P><P>I used&nbsp;<STRONG>proc</STRONG> <STRONG>gee</STRONG>&nbsp;to estimate initial values for parameters.</P><P>&nbsp;</P><P><EM><STRONG>proc</STRONG> <STRONG>gee</STRONG> data=new;</EM></P><P><EM>&nbsp;class var1 id ;</EM></P><P><EM>&nbsp;Model res= var1 var2 var3&nbsp; var4 var5 var6 var7 / dist= mult;</EM></P><P><EM>&nbsp;repeated subject=id / logor=exch;</EM></P><P><EM><STRONG>run</STRONG>;</EM></P><P>&nbsp;</P><P>Also I used these codes to find s1,s2 and s12.</P><P>&nbsp;</P><P><EM>proc means data=new STD;</EM><BR /><EM>var PP LL;</EM><BR /><EM>run;</EM></P><P>&amp;</P><P><EM>proc corr data=new ;</EM><BR /><EM>var PP LL;</EM><BR /><EM>run;</EM></P><P>&nbsp;</P><P>But I have problem with estimate of the s1, s2 and s12 in PROC NLMIXED and receive different errors. The result for my parameters is meaningless.</P><DIV class="branch"><DIV><DIV align="center">Parameter EstimatesParameter Estimate StandardError DF t&nbsp;Value Pr&nbsp;&gt;&nbsp;|t| 95% Confidence Limits Gradienta1a2var1_avar2_avar3_avar4_avar5_avar6_avar7_ab1b2var1_bvar2_bvar3_bvar4_bvar5_bvar6_bvar7_bs1s2s12 <TABLE cellspacing="0" cellpadding="5"><TBODY><TR><TD>-3.4120</TD><TD>1.4995</TD><TD>62</TD><TD>-2.28</TD><TD>0.0264</TD><TD>-6.4095</TD><TD>-0.4145</TD><TD>-54.9813</TD></TR><TR><TD>0.2214</TD><TD>1.1533</TD><TD>62</TD><TD>0.19</TD><TD>0.8484</TD><TD>-2.0840</TD><TD>2.5267</TD><TD>71.2026</TD></TR><TR><TD>-0.03426</TD><TD>0.09628</TD><TD>62</TD><TD>-0.36</TD><TD>0.7232</TD><TD>-0.2267</TD><TD>0.1582</TD><TD>49.4322</TD></TR><TR><TD>0.09149</TD><TD>0.3459</TD><TD>62</TD><TD>0.26</TD><TD>0.7923</TD><TD>-0.5999</TD><TD>0.7828</TD><TD>0.32528</TD></TR><TR><TD>-0.02989</TD><TD>0.06602</TD><TD>62</TD><TD>-0.45</TD><TD>0.6524</TD><TD>-0.1619</TD><TD>0.1021</TD><TD>-91.7194</TD></TR><TR><TD>0.08576</TD><TD>0.04168</TD><TD>62</TD><TD>2.06</TD><TD>0.0438</TD><TD>0.002453</TD><TD>0.1691</TD><TD>50.9831</TD></TR><TR><TD>-0.06421</TD><TD>0.06531</TD><TD>62</TD><TD>-0.98</TD><TD>0.3294</TD><TD>-0.1948</TD><TD>0.06635</TD><TD>72.6902</TD></TR><TR><TD>-0.3454</TD><TD>0.07183</TD><TD>62</TD><TD>-4.81</TD><TD>&lt;.0001</TD><TD>-0.4890</TD><TD>-0.2018</TD><TD>-58.7358</TD></TR><TR><TD>-0.09685</TD><TD>0.08108</TD><TD>62</TD><TD>-1.19</TD><TD>0.2368</TD><TD>-0.2589</TD><TD>0.06523</TD><TD>179.168</TD></TR><TR><TD>-4.5876</TD><TD>118.41</TD><TD>62</TD><TD>-0.04</TD><TD><FONT face="arial black,avant garde">0.9692</FONT></TD><TD>-241.28</TD><TD>232.10</TD><TD>-0.00321</TD></TR><TR><TD>-2.1619</TD><TD>141.76</TD><TD>62</TD><TD>-0.02</TD><TD><FONT face="arial black,avant garde">0.9879</FONT></TD><TD>-285.53</TD><TD>281.21</TD><TD>0.000434</TD></TR><TR><TD>-0.02860</TD><TD>9.8942</TD><TD>62</TD><TD>-0.00</TD><TD><FONT face="arial black,avant garde">0.9977</FONT></TD><TD>-19.8067</TD><TD>19.7495</TD><TD>-0.01680</TD></TR><TR><TD>0.4285</TD><TD>27.2215</TD><TD>62</TD><TD>0.02</TD><TD><FONT face="arial black,avant garde">0.9875</FONT></TD><TD>-53.9864</TD><TD>54.8434</TD><TD>-0.00174</TD></TR><TR><TD>0.1637</TD><TD>11.2263</TD><TD>62</TD><TD>0.01</TD><TD><FONT face="arial black,avant garde">0.9884</FONT></TD><TD>-22.2773</TD><TD>22.6047</TD><TD>-0.01512</TD></TR><TR><TD>0.03240</TD><TD>2.5030</TD><TD>62</TD><TD>0.01</TD><TD><FONT face="arial black,avant garde">0.9897</FONT></TD><TD>-4.9711</TD><TD>5.0359</TD><TD>-0.03055</TD></TR><TR><TD>0.1275</TD><TD>6.2623</TD><TD>62</TD><TD>0.02</TD><TD><FONT face="arial black,avant garde">0.9838</FONT></TD><TD>-12.3907</TD><TD>12.6457</TD><TD>-0.02998</TD></TR><TR><TD>-0.1555</TD><TD>4.8749</TD><TD>62</TD><TD>-0.03</TD><TD><FONT face="arial black,avant garde">0.9747</FONT></TD><TD>-9.9004</TD><TD>9.5894</TD><TD>-0.00437</TD></TR><TR><TD>-0.00080</TD><TD>5.9949</TD><TD>62</TD><TD>-0.00</TD><TD><FONT face="arial black,avant garde">0.9999</FONT></TD><TD>-11.9843</TD><TD>11.9827</TD><TD>-0.03039</TD></TR><TR><TD>0.06539</TD><TD>0.01152</TD><TD>62</TD><TD>5.68</TD><TD>&lt;.0001</TD><TD>0.04237</TD><TD>0.08842</TD><TD>124.931</TD></TR><TR><TD>1.9650</TD><TD>6.7415</TD><TD>62</TD><TD>0.29</TD><TD>0.7717</TD><TD>-11.5111</TD><TD>15.4411</TD><TD>-0.00558</TD></TR><TR><TD>-0.03700</TD><TD>0.01338</TD><TD>62</TD><TD>-2.76</TD><TD>0.0075</TD><TD>-0.06375</TD><TD>-0.01024</TD><TD>-0.00096</TD></TR></TBODY></TABLE></DIV></DIV></DIV><P>As you can see, the p-values for second joint variables are meaningless. So the results do not seem correct.</P><P>How can I find the best s1, s2 and s12&nbsp; for my joint model?&nbsp;</P><P>Thanks for your answer.</P><P>Mahdia</P><P>&nbsp;</P><P>&nbsp;</P> Thu, 21 May 2020 01:28:49 GMT https://communities.sas.com/t5/SAS-Programming/Finding-the-correct-SD-in-PROC-NLMIXED/m-p/649426#M194698 MahdiA110 2020-05-21T01:28:49Z Re: Finding the correct SD in PROC NLMIXED https://communities.sas.com/t5/SAS-Programming/Finding-the-correct-SD-in-PROC-NLMIXED/m-p/649428#M194700 <P>The table is available in attachment</P> Thu, 21 May 2020 01:39:36 GMT https://communities.sas.com/t5/SAS-Programming/Finding-the-correct-SD-in-PROC-NLMIXED/m-p/649428#M194700 MahdiA110 2020-05-21T01:39:36Z Re: Finding the correct SD in PROC NLMIXED https://communities.sas.com/t5/SAS-Programming/Finding-the-correct-SD-in-PROC-NLMIXED/m-p/651586#M195510 No one have any suggestion or idea for me????? Thu, 28 May 2020 23:20:24 GMT https://communities.sas.com/t5/SAS-Programming/Finding-the-correct-SD-in-PROC-NLMIXED/m-p/651586#M195510 MahdiA110 2020-05-28T23:20:24Z