{smcl}
{com}{sf}{ul off}{txt}{.-}
       log:  {res}C:\Documents and Settings\jli2\My Documents\Log\MLE_log\Linear_414.smcl
  {txt}log type:  {res}smcl
 {txt}opened on:  {res}29 Sep 2004, 14:54:40
{txt}
{com}. * linear regression model
. * Wooldridge textbook, page 80, exercise 4.14
. use http://fmwww.bc.edu/ec-p/data/wooldridge/ATTEND
{txt}
{com}. * use "canned command--regress" to run regression
. regress stndfnl atndrte frosh soph

      {txt}Source {c |}       SS       df       MS              Number of obs ={res}     680
{txt}{hline 13}{char +}{hline 30}           F(  3,   676) ={res}    6.74
    {txt}   Model {char |} {res} 19.3023743     3  6.43412478           {txt}Prob > F      = {res} 0.0002
    {txt}Residual {char |} {res} 645.461067   676  .954824063           {txt}R-squared     = {res} 0.0290
{txt}{hline 13}{char +}{hline 30}           Adj R-squared = {res} 0.0247
    {txt}   Total {char |} {res} 664.763441   679   .97903305           {txt}Root MSE      = {res} .97715

{txt}{hline 13}{c TT}{hline 64}
     stndfnl {c |}      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
{hline 13}{char +}{hline 64}
     atndrte {c |}  {res} .0081634   .0022031     3.71   0.000     .0038376    .0124892
{txt}       frosh {c |}  {res}-.2898943   .1157244    -2.51   0.012    -.5171167   -.0626719
{txt}        soph {c |}  {res}-.1184455   .0990267    -1.20   0.232    -.3128824    .0759913
{txt}       _cons {c |}  {res}-.5017308   .1963139    -2.56   0.011    -.8871892   -.1162724
{txt}{hline 13}{c BT}{hline 64}

{com}. * use self-defined maximum likelihood estimator
. program define linear_regress
{txt}  1{com}. args lnf theta1 theta2
{txt}  2{com}. quietly replace `lnf'=-0.5*ln(`theta2')-($ML_y1-`theta1')^2/(2*`theta2')
{txt}  3{com}. end
{txt}
{com}. ml model lf linear_regress ( stndfnl= atndrte frosh soph) ()
{txt}
{com}. ml maximize

{txt}initial:       log likelihood = {res}    -<inf>{txt}  (could not be evaluated)
feasible:      log likelihood = {res}-579.52351
{txt}rescale:       log likelihood = {res}-579.52351
{txt}rescale eq:    log likelihood = {res}-332.38258
{txt}Iteration 0:   log likelihood = {res}-332.38258{txt}  
Iteration 1:   log likelihood = {res}-322.63624{txt}  
Iteration 2:   log likelihood = {res}-322.27735{txt}  
Iteration 3:   log likelihood = {res}-322.27651{txt}  
Iteration 4:   log likelihood = {res}-322.27651{txt}  

{col 51}Number of obs{col 67}= {res}       680
{col 51}{txt}Wald chi2({res}3{txt}){col 67}= {res}     20.34
{txt}Log likelihood = {res}-322.27651{col 51}{txt}Prob > chi2{col 67}= {res}    0.0001

{txt}{hline 13}{c TT}{hline 64}
     stndfnl {c |}      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
{hline 13}{c +}{hline 64}
{res}eq1          {txt}{c |}
     atndrte {c |}  {res} .0081634   .0021966     3.72   0.000     .0038581    .0124687
       {txt}frosh {c |}  {res}-.2898943   .1153835    -2.51   0.012    -.5160418   -.0637468
        {txt}soph {c |}  {res}-.1184455    .098735    -1.20   0.230    -.3119626    .0750715
       {txt}_cons {c |}  {res}-.5017308   .1957357    -2.56   0.010    -.8853657   -.1180959
{txt}{hline 13}{c +}{hline 64}
{res}eq2          {txt}{c |}
       _cons {c |}  {res} .9492075    .051478    18.44   0.000     .8483124    1.050103
{txt}{hline 13}{c BT}{hline 64}

{com}. * in the same fashion, you can try exercise 4.13 on page 80
. log close
       {txt}log:  {res}C:\Documents and Settings\jli2\My Documents\Log\MLE_log\Linear_414.smcl
  {txt}log type:  {res}smcl
 {txt}closed on:  {res}29 Sep 2004, 15:01:56
{txt}{.-}
{smcl}
{txt}{sf}{ul off}