Name

adf — adf unit root test

CALLING SEQUENCE

[result]=adf(namey,p,l,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 = -1, for no added term

 - p = 0, for constant term

 - p = 1, for constant plus time-trend

 - p > 1, for higher order polynomial

• l: # of lags of the ADF test

• arg1,…,argn = optional arguments which can be:

  - 'noprint' if the user doesn't want to print the results of the regression

   - 'dropna' if the user wants to remove the NA values from the data

Output

• resadf = results tlist with:

  - resadf('meth') = 'adf'

  - resadf('y') = y data vector of the auxiliary regression

  - resadf('x') = x data matrix of the auxiliary regression

  - resadf('nobs') = # observations

  - resadf('nvar') = # variables

  - resadf('beta') = bhat

  - resadf('yhat') = yhat

  - resadf('resid') = residuals of the auxiliary regression

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

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

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

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

  - resadf('tstat') = t-stats

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

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

  - resadf('condindex') = multicolinearity cond index

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

  - resadf('rsqr') = rsquared

  - resadf('rbar') = rbar-squared

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

  - resadf('pvaluef') = its significance level

  - resadf('prests') = boolean indicating the presence or absence of a time series in the regression

  - resadf('namey') = name of the y variable of the auxiliary regression

  - resadf('namex') = name of the x variables of the auxiliary regression

  - resadf('bounds') = if there is a timeseries in the regression, the bounds of the regression

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

  - resadf('1% level') = 1% critical level

  - resadf('5% level') = 5% critical level

  - resadf('10% level') = 10% critical level

  - resadf('dropna') = boolean indicating if NAs have been droped

  - resadf('nonna') = vector indicating position of non-NA values (if the option 'dropna' was active)

DESCRIPTION

Computes the augmented Dickey-Fuller test.

EXAMPLE

1) adf('log(inc)',1,4)
2) adf('log(inc)',0,4)
 
Examples taken from function adf_d. Provide adf test for variable log(inc) taken from data base cousa with a simple time trend and a constant respectively.

AUTHOR

Eric Dubois 2002-2007