<< olspec Single equation regressions olspecmod >>
grocer >> Single equation regressions > olspec1

olspec1

ols with specification test

CALLING SEQUENCE

[rols]=olspec1(y,x,names,ltest,test_default)

PARAMETERS

Input

* y = a (n x 1) vector

* x = a (n x k) vector

* names = a tlist, typed 'names' with: anmes('name_foo') = a string, the name that will be given in the corresponding test in the results displayed

* ltest = a string, 'test=xxx' where xxx collects the names of the specification tests performed (for instance 'test=predfailin(0.5),predfailin(0.9), doornhans,arlm(4),hetero_sq')

* test_default = the testing function that will be used by default

Output

* rols = a results tlist with

   - rols('meth') = 'ols'

   - rols('y') = y data vector

   - rols('x') = x data matrix

   - rols('nobs') = # observations

   - rols('nvar') = # variables

   - rols('beta') = bhat

   - rols('yhat') = yhat

   - rols('resid') = residuals

   - rols('vcovar') = estimated variance-covariance matrix of beta

   - rols('sige') = estimated variance of the residuals

   - rols('sigu') = sum of squared residuals

   - rols('ser') = standard error of the regression

   - rols('tstat') = t-stats

   - rols('pvalue') = pvalue of the betas

   - rols('dw') = Durbin-Watson Statistic

   - rols('condindex') = multicolinearity cond index

   - rols('prescte') = boolean indicating the presence or absence of a constant in the regression

   - rols('like') = log-likelihood of the regression

   - rols('rsqr') = rsquared

   - rols('rbar') = rbar-squared

   - rols('f') = F-stat for the nullity of coefficients other than the constant

   - rols('pvaluef') = its significance level

   - rols('name_test') = the names of the specification

   - rols('spec_test') = a matrix with the values of the statistics of the specification tests in column 1 and the corresponding p-values in column 2

DESCRIPTION

Performs ordinary least-squares and standard specification test (Chow in-sample stability tests computed at 50% and 90% of the sample; Doornik and Hansen normality test; heteroskedasticity test called xiª by D.F. Hendry; AutoRegressive Lagrange multiplier tests at order 4) or any tests given by the user.

EXAMPLE

load(GROCERDIR+'data\bdhenderic.dat')
[y,namey,x]=explouniv('delts(lm1-lp)',['delts(lp)','delts(lagts(1,lm1-lp-ly))','rnet','lagts(1,lm1-lp-ly)','cte'],['1964q3';'1989q2'])
names=tlist(['names';'jbnorm';'doornhans';'hetero_sq';...
     'chowtest';'predfailin';'arlm';'reset'],'Jarque and Bera',...
     'Doornik and Hansen','hetero x_squared','Chow',...
     'Chow pred. fail. ','AR','reset');
 
r=olspec1(y,x,names,list(),test_spec0)
// diplay the estimated parameters
r('beta')
// display the specifiaction tests results and the corresponding p-values
// note the olspec1 should be used only in other programs or for testing: olspec is the program to use if you want
// a more user-friendly function
r('spec_test')

AUTHOR

Éric Dubois 2019

Report an issue
<< olspec Single equation regressions olspecmod >>