pd_stationary#
Purpose#
Computes the panel data KPSS test.
Format#
- { testd_hom, testd_het, m_lee_est, brks } = pd_stationary(y, model[, nbreak, bwl, varm, pmax, bCtl])#
- Parameters:
y (TxN matrix) – Wide format panel data.
model (Scalar) –
Optional, Model to be implemented.
1
Constant (Default)
2
Constant and trend
test (String) –
Optional, Test to be conducted.
”st”
Stationary tests, no modifications.
”ca”
Based on CA (cross-section averages approach).
”fourier”
CA approach with smooth breaks (fourier approach).
”panic”
Based on PANIC approach
varm (Scalar) –
Optional, long-run consistent variance estimation method. Default = 1.
1
iid.
2
Bartlett.
3
Quadratic Spectral (QS).
4
SPC with Bartlett (Sul, Phillips & Choi, 2005)
5
SPC with QS
6
Kurozumi with Bartlett
7
Kurozumi with QS
bwl (Scalar) – Optional, bandwidth for the spectral window. Default = round(4 * (T/100)^(2/9)).
kmax (Scalar) – Optional, maximum number of factors. Default = 5.
ic (Scalar) –
Optional, information criterion for optimal number of factors. Default = 1.
1
Akaike.
2
Schwarz.
3
t-stat significance.
- Returns:
Nkpss (Dataframe) – The KPSS statistics for each cross-section and the corresponding p-values.
W (Scalar) – Panel stationarity statistic by Hadri (2000) and the corresponding p-value.
P (Matrix) – Panel stationarity statistic by Yin & Wu (2001) and the corresponding p-value.
Pm (Scalar) – Panel stationarity statistic by Nazlioglu et al. (2021) and the corresponding p-value.
Z (Matrix) – Panel stationarity statistic by Nazlioglu et al. (2021) and the corresponding p-value.
Examples#
new;
cls;
library tspdlib;
// Load date file
y = loadd(getGAUSSHome() $+ "pkgs/tspdlib/examples/pd_full.csv",
". + date($Date, '%b-%y')");
/*
** Classical panel stationarity test
*/
// With constant
model = 1;
{ Nkpss, W, P, Pm, Z} = pd_stationary(y, model);
/*
** Cross-section approach panel stationarity test
*/
// Set test
test = "ca";
// With constant
model = 1;
{ Nkpss, W, P, Pm, Z} = pd_stationary(y, model, test);
Source#
pd_stationary.src