Navigation: Procedures >

ARAUTOLAGS Procedure

 

 

 

 

@ARAutoLags computes an information criterion for various lags of AR processes using the Burg or Yule estimates of the partial autocorrelations. @MAAutoLags does a similar "quick" calculation for choosing a pure MA process. @BJAUTOFIT and @GMAUTOFIT do more time-consuming calculations (they do exact maximum likelihood fits) for choosing full ARMA models. (@GMAUTOFIT is for seasonal ARMA models).

 

@ARAutoLags( options ) series  start  end

Parameters

series

series to analyze

start, end

range of series to use (by default, the maximum possible)

Options

METHOD=[YULE]/BURG

Method used to compute the autocorrelations. YULE is more commonly used (it's the standard "textbook" calculation), but BURG more accurate if the process is nearly non-stationary.

 

MAXLAGS=maximum number of lags to consider [25]

 

CRIT=[AIC]/BIC/CAIC/HQ

Criterion to use:

AIC is (uncorrected) Akaike information criterion.

BIC is the Bayesian or Schwarz criterion.

HQ is Hannan-Quinn.

CAIC is the AIC-corrected for degrees of freedom

 

[PRINT]/NOPRINT

   Show output

 

TABLE/[NOTABLE]

TITLE=title for table ["'criterion' analysis of 'series'"]

Show full table of results (not just best)

Variables Defined

%%AUTOP

number of parameters selected (INTEGER)

Example

*

* Replication file for Bauwens and Laurent(2005), "A New Class of

* Multivariate Skew Densities, With Application to Generalized

* Autoregressive Conditional Heteroscedasticity Models," JBES, vol 23,

* 346-354.

*

* Exchange rate example

*

open data txch.xls

data(format=xls,org=columns,left=2) 1 3065 date eurusd yenusd gbpusd $

  r_eurusd r_yenusd r_gbpusd

*

* The returns series exist from the start of the data set.

*

* The authors use a 1 lag AR for the euro and yen---it's not clear how

* they came to that decision on the yen, which seems to have almost no

* noticeable serial correlation. BIC picks zero for all, but we'll stick

* with the choices in the paper.

*

@arautolags(maxlags=10,crit=bic,table) r_eurusd

@arautolags(maxlags=10,crit=bic,table) r_yenusd

@arautolags(maxlags=10,crit=bic,table) r_gbpusd

 

Sample Output

In the output, small is good. The chosen (starred) model is the AR(0) in each case, though the choice between 0 and 1 for R_EURUSD is a very close one.

 

Schwarz/Bayesian IC Lag Analysis for R_EURUSD

Lags      IC

   0    -0.679*

   1    -0.678

   2    -0.676

   3    -0.673

   4    -0.671

   5    -0.668

   6    -0.666

   7    -0.664

   8    -0.661

   9    -0.659

  10    -0.656


 

Schwarz/Bayesian IC Lag Analysis for R_YENUSD

Lags      IC

   0    -0.531*

   1    -0.528

   2    -0.526

   3    -0.525

   4    -0.522

   5    -0.520

   6    -0.517

   7    -0.515

   8    -0.512

   9    -0.510

  10    -0.507


 

Schwarz/Bayesian IC Lag Analysis for R_GBPUSD

Lags      IC

   0    -0.925*

   1    -0.922

   2    -0.922

   3    -0.919

   4    -0.917

   5    -0.916

   6    -0.914

   7    -0.912

   8    -0.910

   9    -0.907

  10    -0.905


 

 

 

 

Copyright © 2024 Thomas A. Doan