model_estimation: Robust structural change estimation

View source: R/Main.R

model_estimationR Documentation

Robust structural change estimation

Description

Function executes main procedures described in Bai and Perron

Usage

model_estimation(
  y_name,
  z_name = NULL,
  x_name = NULL,
  data,
  eps1 = 0.15,
  m = -1,
  prewhit = 1,
  robust = 1,
  hetdat = 1,
  hetvar = 1,
  hetomega = 1,
  hetq = 1,
  procedure = c("doglob", "dotest", "dospflp1", "doorder", "dosequa", "dorepart"),
  CI = 1,
  method = c("BIC", "LWZ", "seq", "rep", "fix"),
  maxi = 10,
  fixb = 0,
  eps = 1e-05,
  betaini = 0,
  fixn = -1,
  printd = 0
)

Arguments

data

the data set for estimation

m

Maximum number of structural changes allowed. If not specify, m will be set to default value from eps1

prewhit

set to 1 if want to apply AR(1) prewhitening prior to estimating the long run covariance matrix.

robust

set to 1 if want to allow for heterogeneity and autocorrelation in the residuals, 0 otherwise. The method used is Andrews(1991) automatic bandwidth with AR(1) approximation and the quadratic kernel. Note: Do not set to 1 if lagged dependent variables are included as regressors.

hetdat

option for the construction of the F tests. Set to 1 if want to allow different moment matrices of the regressors across segments. If hetdat = 0, the same moment matrices are assumed for each segment and estimated from the ful sample. It is recommended to set hetdat=1 if number of regressors x > 0.

hetvar

option for the construction of the F tests.Set to 1 if want to allow for the variance of the residuals to be different across segments. If hetvar=0, the variance of the residuals is assumed constant across segments and constructed from the full sample. This option is not available when robust=1

hetomega

used in the construction of the confidence intervals for the break dates. If hetomega=0, the long run covariance matrix of zu is assumed identical across segments (the variance of the errors u if robust=0)

hetq

used in the construction of the confidence intervals for the break dates. If hetq=0, the moment matrix of the data is assumed identical across segments

procedure

Selectable procedures to execute:

  • doglob A recursive procedure to obtain global minimizers of SSR, corresponding to the break date.

  • dotest Procedure to conduct SupF test of 0 versus m breaks and Double Max test.

  • dospflp1 Procedure to conduct SupF(l+1|l).

  • doorder Procedure to find number of break by criteria selection

  • dosequa Procedure to obtain break dates via sequential method

  • dorepart Procedure to obtain break dates via repartition method

CI

Option to estimate dates, model parameters and their confidence intervals. If 1, specify method to estimate in method below.

method
  • "BIC" Estimate and compute confidence intervals following order selection method to pick number of breaks by BIC. For more details, see doorder,estim

  • "LWZ" Estimate and compute confidence intervals following order selection method to pick number of breaks by LWZ. For more details, see doorder,estim

  • "seq" Estimate and compute confidence intervals following sequential method estimation of the break date. For more details, see dosequa,estim

  • "rep" Estimate and compute confidence intervals following repartition method estimation of the break date. For more details, see dorepart,estim

  • "fix" Estimate and compute confidence intervals following pre-specified number of breaks, i. For more details, see estim

maxi

number of maximum iterations for recursive calculations of finding global minimizers.default = 10

fixb

Option to use initial values of beta. Set to 1 if initial values are entered by users.

eps

convergence criterion for recursive calculations

betaini

values of initial betas. Only used when fixb = 1.

fixn

number of pre-specified breaks. default = -1. It will be replaced automatically to 2 if no specification is given

printd

Print option for model estimation. default = 0, to suppress outputs

y

name of dependent variable

z

name of independent variables which coefficients are allowed to change across regimes. default is vector of 1 (Mean-shift model)

x

name of independent variables which coefficients are constant across regimes. default is NULL

Details

Function executes the main procedures selected by users. There are 7 main procedures including:

  • doglob A recursive procedure to obtain global minimizers of SSR, corresponding to the break date.

  • dotest Procedure to conduct SupF test of 0 versus m breaks and Double Max test.

  • dosupflp1 Procedure to conduct SupF(l+1|l).

  • doorder Procedure to find number of break by criteria selection

  • dosequa Procedure to obtain break dates via sequential method

  • dorepart Procedure to obtain break dates via repartition method

To obtain the confidence intervals for the break dates and corrected errors for estimates of the model, set est = 1. If est = 1, specify one of the following methods for estimation procedure:

  • "BIC" Estimate and compute confidence intervals following order selection method to pick number of breaks by BIC. For more details, see doorder,estim

  • "LWZ" Estimate and compute confidence intervals following order selection method to pick number of breaks by LWZ. For more details, see doorder,estim

  • "seq" Estimate and compute confidence intervals following sequential method estimation of the break date. For more details, see dosequa,estim

  • "rep" Estimate and compute confidence intervals following repartition method estimation of the break date. For more details, see dorepart,estim

  • "fix" Estimate and compute confidence intervals following pre-specified number of breaks, i. For more details, see estim

All default values of error assumptions (robust, hetdat, hetvar, hetq) will be set to 1


RoDivinity/BP2003 documentation built on Oct. 9, 2022, 9:33 a.m.