<< Moon_Perron Panel unit root tests Pesaran >>

Grocer >> Panel unit root tests > Moon_Perron1

Moon_Perron1

Moon and Perron Panel Unit Root Test

CALLING SEQUENCE

res = Moon_Perron1(Y,t_order,kmax,criteria,bandwidth,kern)

PARAMETERS

Input

* Y = matrix (T,N) of observations

  The data matrix may be unbalanced.

  Missing Values must be specified as %nan

* t_order = -1 : no individual effect

      0 : individual effects (Default)

      1 : individual effects and time trends

* LagOrders =

  - %nan if the user wants the program to determine the optimal lag order for each country (pmax=12 or T/4)

  - a vector (N x 1) or (1 x N) of optimal lags for all the individuals of the panel

* pmax = Maximum of the lag order authorized

* bandwidth = 'C' (Default) common lag troncature for the Bartlett kern (Moon and Perron 2003)

        'N' for the Newey-West (1994)'s non parametric bandwidth parameter

        'A' for the Andrews (1991) automatic bandwidth paramete r selection with AR(1) structure

* kern = 'B' for Bartlett (Default)

      'QS' for Quadratic Spectral (not possible when bandwidth = 'C')

 

Output

res = a results tlist with:

  - res('meth') = 'Levin-Lin'

  - res('y') = (T x k) matrix of data

  - res('t_order') = the trend order (-1, 0 or 1)

  - res('t_orderlit') = the trend order in plain English

  - res('ta_star') = Statistic ta_star

  - res('tb_star') = Statistic tb_star

  - res('ta_pvalue') = Pvalue for the statistics ta_star

  - res('tb_pvalue') = Pvalue for the statistics tb_star

  - res('critical') = Normal Critical Values for the statistics ta_star and tb_star at 1%, 5% and 10%

  - res('rho_pool') = Pooled OLS estimator on initial series

  - res('rho_star') = Modified pooled OLS estimator using de-factored data

  - res('khat') = Estimated numbers of Factor with IC1, IC2, IC3, PC1, PC2, PC3, AIC3 and BIC3

  - res('criteria') = Criteria used to estimate the number of common factors. Default value = 1 (IC1)

  - res('IC') = IC1, IC2 and IC3 Information criterions for r=1,...,rmax

  - res('PC') = PC1, PC2 and PC3 Information criterions for r=1,...,rmax

  - res('BIC3') = BIC3 Information criterion for k=1,...,kmax (only BIC criteria function of N and T)

  - res('AIC3') = AIC3 Information criterion (only AIC criteria function of N and T): it tends to overestimate k

  - res('kmax') = Maximum number of common factors authorized

  - res('h') = Values of individual bandwidth parameters

  - res('LRV') = Estimated of Individual Long Run Variances

  - res('TLRV') = Estimated of Individual Temporal Long Run Variances

  - res('kernel') = kern function used

  - res('bandwidth') = Method to fix the bandwidth parameter

DESCRIPTION

Moon and Perron (2003) Unit Root Tests on Panel Data (ref: "Testing for a Unit Root in Panels with Dynamic Factors", Journal of Econometrics, Elsevier, vol. 122(1), p. 81-126, September).

EXAMPLE

load(GROCERDIR+'/data/gdpan_oecd.dat');
Y=explone(dblist(GROCERDIR+'/data/gdpan_oecd.dat'));
// extract matrix of data present in database
res=Moon_Perron1(log(Y),0,5,'BIC3','n','qs')
// Provides the Moon and Perron test for all (logarithm of) Y columns, with a constant but not trend (arg # 2 = 0),
// the maximum # of factors allowed set to 5 (arg # 3 = 5),
// the number of factors determined by the criteria 'BIC3' proposed by Bai an Ng (arg # 4 = 'BIC3'),
// the Newey-West bandwidth (arg # 5 = 'n'), and the Quadratic Spectral kernel (arg # 6 = qs').
 
res=Moon_Perron1(log(Y),1,5,'AIC1','n','b')
// provides the Moon and Perron test for all (logarithm of) Y columns, with individual constants and trend (arg # 2 = 1),
// the maximum # of factors allowed set to 5 (arg # 3 = 5),
// the number of factors determined by the criteria 'AIC1' proposed by Bai an Ng (arg # 4 = 'AIC1'),
// the Newey-West bandwidth (arg # 5 = 'n'), and the Barlett kernel (arg # 6 = b').

AUTHOR

Christophe Hurlin 2004 / Eric Dubois 2008

Report an issue
<< Moon_Perron Panel unit root tests Pesaran >>