topic proc mixed parameter dispersion matrix in Statistical Procedures https://communities.sas.com/t5/Statistical-Procedures/proc-mixed-parameter-dispersion-matrix/m-p/194009#M10333 <HTML><HEAD></HEAD><BODY><P><SPAN style="color: #000000;">The UCLA HLM site has several examples using the HSB dataset on this page: <A href="http://www.ats.ucla.edu/stat/hlm/seminars/hlm_mlm/608/mlm_hlm_seminar_v608.htm"><SPAN style="color: #000000;">http://www.ats.ucla.edu/stat/hlm/seminars/hlm_mlm/608/mlm_hlm_seminar_v608.htm</SPAN></A></SPAN></P><P><SPAN style="color: #000000;">The third model enters SES as a group-mean centered variable at level-1.&nbsp; HLM output reports a .908 reliability estimate for the intercept and a .260 reliability estimate associated with SES.&nbsp; </SPAN></P><P></P><P><SPAN style="color: #000000;">Using the variance components reported for the intercept and the level-1 residual, I’m able to calculate a lambda value for each schoolid (n=160), and then sum across those to obtain an overall reliability estimate that matches the HLM output.&nbsp; However, I’ve been unable to do similar calculations and arrive at the .260 provided for the SES reliability.&nbsp; </SPAN></P><P></P><P><SPAN style="color: #000000;">It's not clear to me how I can obtain the parameter dispersion matrix which would contain both parameter and error dispersion around each Beta.&nbsp; Essentially, i want to know these two facets about the slope parameter associated with SES.&nbsp; Is there a way to obtain this using ODS?&nbsp; Is there something that can be obtained and further processed inside IML? </SPAN></P><P></P><P><SPAN style="color: #000000;">Any thoughts or insights are greatly appreciated.</SPAN></P><P></P><P><SPAN style="color: #000000;">Jason</SPAN></P></BODY></HTML> Sun, 12 Apr 2015 12:47:19 GMT jsberger 2015-04-12T12:47:19Z proc mixed parameter dispersion matrix https://communities.sas.com/t5/Statistical-Procedures/proc-mixed-parameter-dispersion-matrix/m-p/194009#M10333 <HTML><HEAD></HEAD><BODY><P><SPAN style="color: #000000;">The UCLA HLM site has several examples using the HSB dataset on this page: <A href="http://www.ats.ucla.edu/stat/hlm/seminars/hlm_mlm/608/mlm_hlm_seminar_v608.htm"><SPAN style="color: #000000;">http://www.ats.ucla.edu/stat/hlm/seminars/hlm_mlm/608/mlm_hlm_seminar_v608.htm</SPAN></A></SPAN></P><P><SPAN style="color: #000000;">The third model enters SES as a group-mean centered variable at level-1.&nbsp; HLM output reports a .908 reliability estimate for the intercept and a .260 reliability estimate associated with SES.&nbsp; </SPAN></P><P></P><P><SPAN style="color: #000000;">Using the variance components reported for the intercept and the level-1 residual, I’m able to calculate a lambda value for each schoolid (n=160), and then sum across those to obtain an overall reliability estimate that matches the HLM output.&nbsp; However, I’ve been unable to do similar calculations and arrive at the .260 provided for the SES reliability.&nbsp; </SPAN></P><P></P><P><SPAN style="color: #000000;">It's not clear to me how I can obtain the parameter dispersion matrix which would contain both parameter and error dispersion around each Beta.&nbsp; Essentially, i want to know these two facets about the slope parameter associated with SES.&nbsp; Is there a way to obtain this using ODS?&nbsp; Is there something that can be obtained and further processed inside IML? </SPAN></P><P></P><P><SPAN style="color: #000000;">Any thoughts or insights are greatly appreciated.</SPAN></P><P></P><P><SPAN style="color: #000000;">Jason</SPAN></P></BODY></HTML> Sun, 12 Apr 2015 12:47:19 GMT https://communities.sas.com/t5/Statistical-Procedures/proc-mixed-parameter-dispersion-matrix/m-p/194009#M10333 jsberger 2015-04-12T12:47:19Z