I am running a mixed model for a continuous outcome using JMP Pro 17. In the output for the fixed effects parameter estimates, I have a 5-class categorical disease and the report shows the result for 4 of the 5 levels. How do I get or derive the Std Error for the parameter that is not displayed in the output report? (see the last row in the output copied and pasted below)
Fixed Effects Parameter Estimates | |||||||
Term | Estimate | Std Error | DFDen | t Ratio | Prob>|t| | 95% Lower | 95% Upper |
Intercept | 1022.568 | 168.9938 | 296.6 | 6.05 | <.0001 | 689.9884 | 1355.147 |
305MEp | 0.320214 | 0.006541 | 105107 | 48.95 | <.0001 | 0.307393 | 0.333034 |
PrevLgp[1] | -124.385 | 13.32405 | 102630 | -9.34 | <.0001 | -150.5 | -98.2698 |
PrevLgp[2] | 99.0082 | 14.12753 | 90441 | 7.01 | <.0001 | 71.31838 | 126.698 |
PrevLgp[3] | 153.9814 | 15.93062 | 94837 | 9.67 | <.0001 | 122.7575 | 185.2052 |
PrevLgp[4] | 81.06569 | 22.15344 | 100970 | 3.66 | 0.0003 | 37.64522 | 124.4862 |
PrevLgp_5+ |
OK - I figured out a work around. In order to get the missing details, I changed the ordering of the categorical variables and then reran the model. Now, I can copy in the previously missing line because the referent population has changed.
Possibly see here:
Yes, I use the Expanded Estimates report when I run least sq means models/ANOVA but I don't think it is an option within the mixed models section. I am running a mixed model with 2 variables listed as random effects and others listed as fixed effects. However, I can't figure out how to get the std error/confidence interval for the unreported category.
OK - I figured out a work around. In order to get the missing details, I changed the ordering of the categorical variables and then reran the model. Now, I can copy in the previously missing line because the referent population has changed.