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. Migration and employment status during the turbulent nineties in Sweden
2. Feature type effects in semantic memory: An event related potentials study
3. The name is absent
4. LABOR POLICY AND THE OVER-ALL ECONOMY
5. The name is absent
6. Non Linear Contracting and Endogenous Buyer Power between Manufacturers and Retailers: Empirical Evidence on Food Retailing in France
7. An Incentive System for Salmonella Control in the Pork Supply Chain
8. A Rational Analysis of Alternating Search and Reflection Strategies in Problem Solving
9. The name is absent
10. TOWARD CULTURAL ONCOLOGY: THE EVOLUTIONARY INFORMATION DYNAMICS OF CANCER