gls_1break#

Purpose#

Computes the GLS unit root test with one structural break.

Format#

{ ADFGLS_min, p, PT_min, MZa_min, MZt_min, MSB_min, MPT_min, s2, tb1_min, lambda, cvPT, cvMPT, cvMZt, cvDFGLS, cvMSB, cvMZa } = gls_1break(y, model[, pmax, ic])#
Parameters:
  • y (Nx1 matrix) – Time series data to be tested.

  • model (Scalar) –

    Model to be implemented.

    0

    Level shift.

    1

    Trend shift.

    2

    Level and trend shift.

  • pmax (Scalar) – Optional, the maximum number of lags for \(\Delta y\). Default = 8.

  • ic (Scalar) –

    Optional, the information criterion used for choosing lags. Default = 3.

    1

    Akaike.

    2

    Schwarz.

    3

    t-stat significance.

Returns:
  • GLStau (Scalar) – ADF-GLS statistic.

  • p (Scalar) – Number of lags for ADF-GLS statistic.

  • PT (Scalar) – PT statistic.

  • Mza (Scalar) – MZalpha statistic.

  • Mzt (Scalar) – MZt statistic.

  • MSB (Scalar) – MSB statistic.

  • MPT (Scalar) – MPT statistic.

  • s2 – Long-run variance based on autoregressive spectral estimate.

  • tb (Scalar) – Location of the break.

  • lambda (Scalar) – Fraction of break (tb/T)

  • cvPT (Vector) – 1%, 5%, and 10% critical values for the PT statistic.

  • cvMza (Vector) – 1%, 5%, and 10% critical values for the MZalpha statistic.

  • cvMzt (Vector) – 1%, 5%, and 10% critical values for the MZt statistic.

  • cvMSB (Vector) – 1%, 5%, and 10% critical values for the MSB statistic.

  • cvMPT (Vector) – 1%, 5%, and 10% critical values for the MPT statistic.

Examples#

library tspdlib;

// Load date file
y = loadd(getGAUSSHome() $+ "pkgs/tspdlib/examples/ts_examples.csv",
                           "Y + date($Date, '%b-%y')");

// Break in level
model = 1;
call gls_1break(y, model);

// Break in level and trend
model = 2;
call gls_1break(y, model);

Source#

gls_1br.src

See also

Functions gls_2breaks()