STABTEST Procedure

@STABTEST computes the Hansen (Nyblom-Hansen) stability test for a linear regression. It computes both individual and joint test statistics from Hansen(1992). This includes a test for the stability of the variance as well. @FLUX is a similar test which applies to non-linear models.

@STABTEST( options ) depvar start end

# list of explanatory variables(in regression format)

Parameters

depvar	dependent variable
start, end	range for regression. By default, the maximum range permitted by all variables involved in the regression.

Options

SMPL=standard SMPL option[not used]

ORDER=series with order in which data are to be added in testing [entry/time order]

[PRINT]/NOPRINT

Controls whether or not the regression and stability test results are printed

TITLE="title for report" ["Hansen Stability Test"]

Variables Defined

%HJOINT	Joint stability test statistic (REAL)
%HSTATS	VECTOR of individual test statistics (for each coefficient then the variance)

Example

* Johnston & DiNardo, Econometric Methods, 4th edition

* Numerical Illustration from section 4.4

* pp 121-126

open data auto1.asc

cal(q) 1959:1

data(format=prn,org=columns) 1959:1 1992:1 gcng gcngq gdc gydq pop16 x2 x3 y

* Figure 4.2 appears to show the series standardized, by subtracting the

* sample mean and dividing by the sample standard deviation. You can do

* this transformation with the instruction DIFF(STANDARDIZE).

diff(standardize) y 1959:1 1973:3 ys

diff(standardize) x3 1959:1 1973:3 x3s

diff(standardize) x2 1959:1 1973:3 x2s

graph(key=attached,klabels=||"Gas","Price","Income"||,$

footer="Figure 4.2 Gasoline Consumption, Income and Price") 3

# ys

# x2s

# x3s

* The StabTest procedure does the Hansen test for parameter instability.

@stabtest y 1959:1 1971:3

# constant x2 x3

Sample Output

@STABTEST includes the estimation of the linear regression. The test information follows the standard LINREG output. In this case, the null hypothesis of stability is very strongly rejected, with stability rejected for the joint test and each individual coefficient.

Linear Regression - Estimation by Least Squares

Dependent Variable Y

Quarterly Data From 1959:01 To 1971:03

Usable Observations 51

Degrees of Freedom 48

Centered R^2 0.9628561

R-Bar^2 0.9613084

Uncentered R^2 0.9999918

Mean of Dependent Variable -7.871992009

Std Error of Dependent Variable 0.118004981

Standard Error of Estimate 0.023211794

Sum of Squared Residuals 0.0258617954

Regression F(2,48) 622.1350

Significance Level of F 0.0000000

Log Likelihood 121.0979

Durbin-Watson Statistic 0.3162

Variable Coeff Std Error T-Stat Signif

************************************************************************************

1. Constant -1.104123750 0.470426008 -2.34707 0.02309462

2. X2 -0.662340328 0.145458781 -4.55346 0.00003620

3. X3 0.847907643 0.060386831 14.04127 0.00000000

Hansen Stability Test

Test Statistic P-Value

Joint 4.06952983 0.00

Variance 0.12107450 0.47

Constant 0.92208827 0.00

X2 0.91741025 0.00

X3 0.91379801 0.00