<< olsecm Unit roots and cointegration rztcrit >>

Grocer >> Unit roots and cointegration > phil_perr

phil_perr

Phillips-Perron unit-root test

CALLING SEQUENCE

[result]=phil_perr(namey,p,l,np,arg1,...,argn)

PARAMETERS

Input

* namey = a time series, a real (nx1) vector or a string equal to the name of a time series or a (nx1) real vector between quotes

* p = order of time polynomial in the null-hypothesis

  -p =  0, for constant term

  -p =  1, for constant plus time-trend

* l= (optional) truncation lag of the Newey-West window default: l = floor(5*nobs^0.25)

 

Output

* result tlist with:

  - result('meth') = 'phillips-perron'

  - result('y') = y data vector

  - result('x') = x data matrix

  - result('nobs') = nobs

  - result('nvar') = nvars

  - result('beta') = bhat

  - result('yhat') = yhat

  - result('resid') = residuals

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

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

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

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

  - result('tstat') = t-stats

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

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

  - result('condindex') = multicolinearity cond index

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

  - result('rsqr') = rsquared

  - result('rbar') = rbar-squared

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

  - result('pvaluef') = its significance level

  - result('zalpha') = Phillips-Peron statistic, autocorrelation/heteroskedasticity corrected value for the unit-root coefficient based on a Dickey-Fuller regression

  - result('zt') = Phillips-Peron statistic, autocorrelation/heteroskedasticity corrected t-ratio for the unit-root coefficient based on a Dickey-Fuller regression

  - result('crit_al') = (3 x 1) vector of critical values [1% 5% 10%] quintiles for zalpha

  - result('crit_t') = (3 x 1) vector of critical values [1% 5% 10%] quintiles for zt

DESCRIPTION

Computes Phillips-Perron test of the unit-root hypothesis based on a Dickey-Fuller/ Augmented Dickey-Fuller regression.

EXAMPLE

load(GROCERDIR+'/data/bdhenderic.dat');
bounds('1964q3','1989q2');
r=phil_perr('lm1',0,15)
r=phil_perr('lm1',1)
 
// Examples taken from fucntion phil_perr_d.
// Example 1 provides the Phillips-Peron test for variable lm1, whitout trend and with a 15 length for the Newey-West window.
// Example 2 provides the Phillips-Peron test for variable lm1, whit trend and with a default length for the Newey-West window.

AUTHOR

Eric Dubois 2002

Report an issue
<< olsecm Unit roots and cointegration rztcrit >>