Question

show all work include any Stata work

You estimate the regression model on the next page in order to study the determinants of wealth. Specifically, data from 9,275 individuals on the following variables are gathered:

            netfai =   net financial wealth, in thousands of dollars, for individual i.

            inci =    income, in thousands of dollars, for individual i.

            agei =     age of individual i

            marriedi =   dummy equal to 1 if individual i is married

            malei =     dummy equal to 1 if individual i is male

           p401ki =   dummy equal to 1 if individual i participates in a 401k retirement plan

           

Interpret the coefficients of age and married. Are they significant? At what level? (Use the p-value)

How much will a 10% increase in average income change wage?

What econometric problems appear to exist and not exist? Be specific, backing up your answers with evidence from the regression and tests in the next page.

Based on the regression results, do males have greater wealth than females? If so, by how much?

USE THE FOLLOWING DATA

. regress nettfa inc age marr male p401k

      Source |       SS       df       MS              Number of obs =    9275

-------------+------------------------------           F( 5, 9269) = 411.56

       Model | 6893418.45     5 1378683.69           Prob > F      = 0.0000

    Residual |    31049971 9269 3349.87281           R-squared     = 0.1817

-------------+------------------------------           Adj R-squared = 0.1812

       Total | 37943389.5 9274   4091.3726           Root MSE      = 57.878

------------------------------------------------------------------------------

      nettfa |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]

-------------+----------------------------------------------------------------

         inc |   .9565831   .0279059    34.28   0.000     .9018815   1.011285

         age |   1.044085   .0590782    17.67   0.000     .9282784    1.159891

        marr | -9.521908   1.433179    -6.64   0.000    -12.33125   -6.712562

        male |   .5348946   1.614591     0.33   0.740    -2.630058    3.699848

       p401k |   13.11685   1.396504     9.39   0.000     10.37939     15.8543

       _cons | -59.11724   2.746083   -21.53   0.000    -64.50017   -53.73431

------------------------------------------------------------------------------

correlate inc age marr male p401k

(obs=9275)

             |      inc      age     marr     male    p401k

-------------+---------------------------------------------

         inc |   1.0000

         age |   0.1056   1.0000

        marr |   0.3620   0.0590   1.0000

        male | -0.0699 -0.1203 -0.3640   1.0000

       p401k |   0.2708   0.0260   0.0856 -0.0249   1.0000

. estat ovtest

Ramsey RESET test using powers of the fitted values of nettfa

       Ho: model has no omitted variables

                F(3, 9266) =    246.86

                  Prob > F =      0.0000

. summarize nettfa inc age marr male p401k

    Variable |       Obs        Mean    Std. Dev.       Min        Max

-------------+--------------------------------------------------------

      nettfa |      9275    19.07168    63.96384   -502.302   1536.798

         inc |      9275    39.25464       24.09     10.008    199.041

         age |      9275    41.08022    10.29952         25         64

        marr |      9275    .6285714    .4832128          0          1

        male |      9275    .2044205    .4032993          0          1

-------------+--------------------------------------------------------

       p401k |      9275    .2762264    .4471543          0          1

Answer 1

There are three questions. As the student didn't mention, I have answered only the first question in detail.

Answer 1.

• The coefficient age is 1.044085 or approx 1.04 which means for each one year increase in age, the nettfa (i.e., the depended variable) increases by approximately 1.04 unit. The coefficient of married is -9.521908 or approx -9.52, which implies that being married we expect a 9.52 unit decrease in the nettfa (i.e., the depended variable).
• Both, age and marital status are significant.
• Two-tail p-value tests the hypothesis - whether the coefficient is different from zero. As the p-values for both the variables (age and marr) are 0.00, the variables are significant at 1% level of significance.
• There is no specification of wage in the estimated regression model.

Answer 2.

• The econometric problems are multicolinearity (due to the high correlation (more than 0.30) between inc and marr) and omitted variables (because null hypothesis in 'ovtest' is rejected).

Answer 3.

• The gender difference is insignificant (as p-value is 0.74) in the estimated model. So, we cannot conclude that males have significantly more wealth than females for the estimated model in the given sample.