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);

Source

pd_iltlevel.src

See also

Functions pd_kpss(), pdfzk()