cips#

Purpose#

Computes a panel data unit root test in the presence of cross-section dependence.

Format#

{ Ncadf, Nmcadf, Nlags, pcadf, pmcadf } = CIPS(y, model[, pmax, ic])#
Parameters:
  • y (TxN matrix) – Wide panel data set to be tested.

  • model (Scalar) –

    Model to be implemented.

    0

    None.

    1

    Constant.

    2

    Constant and trend.

  • 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:
  • Ncadf (Nx1 Vector) – CADF statistics for each cross-section.

  • Nmcadf (Nx1 Vector) – Modified CADF statistics for each cross-section.

  • Nlags (Nx1 Vector) – Number of lags selected by specified information criterion for each cross-section.

  • pcadf (Scalar) – Panel CIPS statistic

  • pmcadf (Scalar) – Modified panel CIPS statistic

Examples#

library tspdlib;

// CADF and Modified CADF tests
/*
** Using the defaults
** for maximum number of lags
** and information criterions,
*/

// Set up model
model = 1;
{ Ncadf, Nlm, Nmcadf, Nlags, pcadf, pmcadf } = cips(y, model);

Source#

pd_cips.src