pdlm¶
Purpose¶
Computes panel data LM tests with 0, 1, or 2 breaks in the level or level and trend.
Format¶
-
{ Nlm, Ntb, Np, PD_lm, cvPDLM } =
PDLM(y, model[, nbreak, pmax, ic, trimm])¶ Parameters: - y (Nx1 matrix) – Time series data to be tested.
- model (Scalar) –
Model to be implemented. Default = 0.
1 Level break model. 2 Level and trend break model. - nbreak (Scalar) –
Optional, number of breaks.
0 No break test. 1 One break test. 2 Two break tests. Default = 0.
- pmax (Scalar) – Optional, maximum number of lags. Default = 8.
- ic (Scalar) –
Optional, the information criterion used for choosing lags. Default = 3.
1 Akaike. 2 Schwarz. 3 t-stat significance. - trimm (Scalar) – Optional, trimming rate. Default = 0.10.
Returns: - Nlm (Vector) – Minimum test statistic for each cross-section.
- Ntb (Vector) – Location of break(s) for each cross-section.
- Np (Vector) – Number of lags selected by choosen information criterion for each cross-section.
- PD_lm (Scalar) – Panel LM statistic with N(0,1).
- pval (Scalar) – P-value of the panel LM statistic.
Examples¶
library tspdlib;
// Load date file
y = loadd(getGAUSSHome() $+ "pkgs/tspdlib/examples/pd_full.csv",
". + date($Date, '%b-%y')");
// Run panel LM test with level shifts
// No breaks
model = 1;
// No breaks
{ Nlm, Ntb, Np, PD_lm, pval } = PDLM(y, model);