lm_1break

Purpose

Computes the minimum LM unit root test with one structural break.

Format

{ lm_tstat, tb, lags, lambda, cv } = LM_1break(y, model[, pmax, ic, trimm])
Parameters:
  • y (Nx1 matrix) – Time series data to be test.
  • model (Scalar) –

    Model to be implemented.

    1 Break in level.
    2 Break in level and trend.
  • pmax (Scalar) – Optional, the maximum number of lags for \(\delta y\). Default = 8.
  • ic – Optional, the information criterion used for choosing lags. Default = 3.
Parameters:

trimm (Scalar) – Optional, trimming rate. Default = 0.10.

Returns:
  • lm_tstat (Scalar) – Minimum test statistic.
  • tb (Scalar) – Location of break.
  • lags (Scalar) – Number of lags selected by chosen information criterion.
  • lambda (Scalar) – Break fraction (tb/T).
  • cv (Vector) – 1, 5, and 10 percent critical values for lm_1break() tau-stat.

Examples

new;
cls;
library tspdlib;

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

// Break in level
model = 1;
{ LM_min, tb1, p, lambda, cv } = LM_1break(y, model);

// Break in level and trend
model = 2;
{ LM_min, tb1, p, lambda, cv } = LM_1break(y, model);

Source

lm_1br.src

See also

Functions lmkpss(), lm_2break()