st: Graphing Growth Curve Model

[Brendan Churchill <Brendan.Churchill@utas.edu.au>]

 

Dear Statalist Users

 

Could you please help me out! I’m trying to graph this growth curve model, but I’m having trouble executing the right command. I’m interesting in graphing the estimations from this model across the  ‘cohort’ variable. Previously I was able to do this successfully when there were fewer variables in the model using the predict command and then specifying ‘predict yhat age if cohort==1’ etc. This produced a nice graph of cohort across age of the predicted variable atwkwrl. When I try this, the graph looks crazy! Any suggestions! I would greatly appreciate any help anyone can offer!

                               

atwkwrl                                Coef.                     Std. Err.      z    P>z             [95% Conf. Interval]

                               

agecohortmedian             .0621632             .0249317     2.49   0.013   .0132979    .1110284

agemediansq                      -.0003774            .0000616    -6.13   0.000   -.0004981   -.0002567

agexcohort                         -.0030805             .0002556   -12.05   0.000 -.0035814   -.0025795

cohort                                   .2677652              .1255535     2.13   0.033   .0216849    .5138455

sexc                                        .2037297             .0280607     7.26   0.000   .1487317    .2587277

_Ieduhistor_2                   .4122904              .0283585    14.54   0.000  .3567088    .4678721

_Ieduhistor_3                   .490523                 .0793989     6.18   0.000   .3349039     .646142

_Imaritalst_2                     -.3290389             .0397241    -8.28   0.000   -.4068968   -.2511811

_Imaritalst_3                     -.0141621             .0334132    -0.42   0.672   -.0796508    .0513267

_Imaritalst_4                     -.4779623             .0405091   -11.80   0.000 -.5573587    -.398566

wagesc                                 -.0001603             .0000222    -7.20   0.000   -.0002039   -.0001167

_Ilfstatus_2                        .0761665              .0309167     2.46   0.014   .0155708    .1367622

_Ilfstatus_3                        .1250215              .0649347     1.93   0.054   -.0022483    .2522912

_Ilfstatus_4                        .108923                 .0360847     3.02   0.003   .0381982    .1796478

_cons                                    1.159001              2.431964     0.48   0.634   -3.607561    5.925562

                               

 

                               

Random-effects Parameters      Estimate   Std. Err.           [95% Conf. Interval]

                               

pid: Unstructured           

var(agecoh~n)  .0000948   .0000151          .0000694    .0001294

var(_cons)          2.149805   .1154502          1.935029    2.388421

cov(agecoh~n,_cons)    -.0142736   .0015029        -.0172193   -.0113279

                               

var(Residual)     1.576647   .0177357          1.542266    1.611794

                               

LR test vs. linear regression:        chi2(3) =  3418.03             Prob > chi2 = 0.0000