<< dfbeta Econometric tests and diagnostics doornhans >>

Grocer >> Econometric tests and diagnostics > diebmar

diebmar

Diebold and Mariano test for predictive accuracy

CALLING SEQUENCE

rdb = diebmar(namey,arg1,...,argn)

PARAMETERS

Input

* namey = observed vector TS: 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

* argi = arguments which can be:

  - 2 time series (the first serie is the benchmark model)

  - a real (nx2) vector

  - a real (nx2) matrix

  - a string equal to the name of a time series or a (nx2) real vector or matrix between quotes

  - a list of such elements

  - 'smallc =1' if the user want to perform a small sample correction will be done (optional; default = 0: no small sample correction)

  - 'trunc = xx' truncation point of the bartlett window (optional; default = floor(5*nobs^0.25))

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

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

 

Output

* rdb = a results tlist with

  - rdb('meth') = 'diebmar'

  - rdb('y') = input forecasted series

  - rdb('x') = forecasts resp. under the null hypothesis and forecast under the alternative hypothesis

  - rdb('namey') = name of forecasted serie

  - rdb('namex') = names of forecasts resp. under the null hypothesis and under the alternative hypothesis

  - rdb('mse') = vector of MSE

  - rdb('stat') = value of the Diebold-Mariano statistics

  - rdb('pvalue') = p-value of the Diebold-Mariano statistics

  - rdb('smallc') = flag for a small sample correction

  - rdb('trunc') = truncation lag of the Barlett window

  - rdb('prests') = flag for the presence of time series

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

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

  - rdb('nonna') = vector indicating position of non-NAs (if the option 'dropna' was active)

DESCRIPTION

A function that computes Diebold and Mariano test for predictive accuracy while allowing for small sample correction as proposed by Harvey et al.

EXAMPLE

bounds('1979q1','2003q4');
ri = rolreg('delts(log(y))','ypast','delts(yfut)','cte',...
'dates=[''1992q1'',''2003q4'']','hstep=1');
rb = rolreg('delts(log(y))','delts(lagts(log(y)))','delts(lagts(2,log(y)))','cte',...
'dates=[''1992q1'',''2003q4'']','hstep=1');
bounds();
 
y_indic =ri('yfor');
y_bench = rb('yfor');
 
bounds('1992q2','2003q4');     // select an appropriate sample period
 
rdb=diebmar('delts(log(y))','y_bench','y_indic','smallc=1','trunc=4")
//Examples taken from function diebmar_d(). The forecasted series is delts(log(y)), the benchmark series is y_bench, the alternative
one is y_indic. Here, a small sample correction is performed and the Barlett window is truncated at 4 lags.
rdb=diebmar('delts(log(y))','y_bench','y_indic','smallc=1')
// Now, a small sample correction is performed and the Barlett window is automatically determined
rdb=diebmar('delts(log(y))','y_bench','y_indic')
// Now, no small sample correction is performed and the Barlett window is the same as in example 2.

AUTHOR

Emmanuel Michaux/ Eric Dubois 2004-2007

Report an issue
<< dfbeta Econometric tests and diagnostics doornhans >>