Bias_Correc_VAR function

Estimates an unbiased VAR(1) using stochastic approximation (Bauer, Rudebusch and Wu, 2012)

Bias_Correc_VAR(
  ModelType,
  BRWinputs,
  RiskFactors,
  N,
  Economies,
  FactorLabels,
  GVARinputs = NULL,
  JLLinputs = NULL,
  ev_restr = 1,
  nargout = 4
)

Arguments

• ModelType: A character vector indicating the model type to be estimated.

• BRWinputs: A list containing the necessary inputs for the BRW model estimation:

  1. flag_mean: Logical. Determines whether mean- (TRUE) or median- (FALSE) unbiased estimation is desired. Default is TRUE.
  2. gamma: Numeric. Adjustment parameter between 0 and 1. Default is 0.5.
  3. N_iter: Integer. Number of iterations for the stochastic approximation algorithm after burn-in. Default is 5,000.
  4. N_burn: Integer. Number of burn-in iterations. Default is 15
  5. B: Integer. Number of bootstrap samples per iteration for calculating the noisy measure of the OLS estimator's mean or median. Default is 50.
  6. check: Logical. Indicates whether to perform a closeness check. Default is TRUE.
  7. B_check: Integer. Number of bootstrap samples for the closeness check. Default is 100,000.

• RiskFactors: A numeric matrix (T x F) representing the time series of risk factors.

• N: Integer. Number of country-specific spanned factors.

• Economies: A character vector containing the names of the economies included in the system.

• FactorLabels: A list of character vectors with labels for all variables in the model.

• GVARinputs: List. Inputs for GVAR model estimation (see GVAR function). Default is NULL.

• JLLinputs: List. Inputs for JLL model estimation (see JLL function). Default is NULL.

• ev_restr: Numeric. Restriction on the largest eigenvalue under the P-measure. Default is 1.

• nargout: Integer. Number of elements in the output list. Default is 4.

Returns

Bias-corrected VAR parameters based on the framework of Bauer, Rudebusch and Wu (2012). The list contains:

1. Phi_tilde: estimated coefficient matrix (F x F);
2. mu_tilde: estimated intercept (F x 1);
3. V_tilde: estimated variance-covariance matrix (F x F);
4. dist: root mean square distance (scalar);
5. Phi_sample: sample estimated variance-covariance matrix used in the checks (F x F x B_check) - this output is reported if nargout is 5.

Examples

data(CM_Factors)
Factors <- t(RiskFactors[1:7,])

BRWinputs <- list(flag_mean = TRUE, gamma = 0.4, N_iter = 1000, N_burn = 100,
                  B = 10, check = 1, B_check = 5000)

Economies <- "China"
N <- 3
ModelType <- "JPS original"
FactorLabels <- NULL

BRWpara <- Bias_Correc_VAR(ModelType, BRWinputs, Factors, N, Economies, FactorLabels)

References

Bauer, Rudebusch and, Wu (2012). "Correcting Estimation Bias in Dynamic Term Structure Models"

This function is based on the est_unb_var Matlab function available at Cynthia Wu's website (https://sites.google.com/view/jingcynthiawu/).