The name is absent



18


Stata Technical Bulletin


STB-33


. eq priv inc yrs

. eq vote inc ptax

. suprob priv vote

Fitting constant only model

Iteration 0: Log Likelihood = -82.529057

Iteration 1: Log Likelihood = -82.078668

Iteration 2: Log Likelihood = -82.076956

Iteration 3: Log Likelihood = -82.076955

Fitting full model

Iteration 0: Log Likelihood = -75.29544

Iteration 1: Log Likelihood = -74.350968

Iteration 2: Log Likelihood = -74.333447

Iteration 3: Log Likelihood = -74.333444

Seemingly unrelated probit regression               Number of obs =      80

Model chi2(4)    =   15.49

Prob > chi2      = 0.0038

Log Likelihood =    -74.3334444                     Pseudo R2        = 0.0943

I

Coef.

Std. Err.

z

P>z

[957. Conf.

Interval]

priv
inc

I
I

.3067012

.4499467

0.682

0.495

-.5751781

1.18858

yrs

I

-.0161475

.0264445

-0.611

0.541

-.0679777

.0356827

-Cons

I

-4.091401

4.569771

-0.895

0.371

-13.04799

4.865184

vote
inc

I
I

1.651935

.5529672

2.987

0.003

.5681397

2.735731

pt ax

I

-2.028817

.7238308

-2.803

0.005

-3.447499

-.6101343

„cons

I

-2.007338

4.075971

-0.492

0.622

-9.996095

5.981419

rho

„cons

I
I

-.3252008

.2240436

-1.452

0.147

-.7643183

.1139166

Example: Robust bivariate probit regression

In this example, we will use the automobile dataset that ships with Stata. We have one binary variable foreign that
denotes whether a car is domestic (foreign = 0) or foreign (foreign = 1). We will also assume for the sake of this example,
that there is another variable guzzler that denotes whether a car is a gas guzzler (guzzler = 1) or not (guzzler = 0). The
guzzler variable was created using gen guzzler = (mpg>=24).

Knowing that most foreign cars imported are smaller and that smaller cars usually get better mileage, we wish to model
these variables with the length and weight of the car.

. biprob foreign guzzler length weight, robust nolog

Bivariate probit regression


Log Likelihood =   -46.7432695


Number of obs    =      74

Model chi2(4)    =   80.56

Prob > chi2      =  0.0000

Pseudo R2        =  0.4629

I                 Robust

I

Coef.

Std. Err.

z

P>z

[957. Conf.

Interval]

foreign
length
weight

_cons

I
I

I
I

.0051157

-.0016416

3.111534

.0272459

.0009073

2.754162

0.188

-1.809

1.130

0.851

0.070

0.259

-.0482852

-.0034198

-2.286524

.0585166

.0001366

8.509591

guzzler
length
weight

_cons

I
I

I
I

-.0622867

-.0008044

12.87024

.0298606

.0008508

3.447198

-2.086

-0.945

3.734

0.037

0.344

0.000

.0037609

-.0008631

-19.62662

.1208124

.002472

-6.113856

rho

_cons

I
I

-.5294745

.2323637

-2.279

0.023

.07405

.984899



More intriguing information

1. The name is absent
2. The name is absent
3. The name is absent
4. Human Resource Management Practices and Wage Dispersion in U.S. Establishments
5. Transfer from primary school to secondary school
6. Gender and headship in the twenty-first century
7. Testing Hypotheses in an I(2) Model with Applications to the Persistent Long Swings in the Dmk/$ Rate
8. Update to a program for saving a model fit as a dataset
9. Testing Gribat´s Law Across Regions. Evidence from Spain.
10. The name is absent