ardl.nardl (version 1.3.0)

gets_ardl_uecm: General-to-specific approach for the autoregressive distributed lag model

Description

Adopt the general-to-specific approach to estimate the autoregressive distributed lag model

Usage

gets_ardl_uecm(x, dep_var, expl_var, p_order = c(2), q_order = c(3),
              gets_pval = 0.1, case = 3, F_HC = FALSE, order_l = 5, 
              graph_save = FALSE)

Value

Parsimonious_ARDL_fit

Return an estimated general-to-specific ARDL model

Parsimonious_ECM_fit

Return an estimated general-to-specific error correction model

Summary_ecm_fit

Return the summary of 'Parsimonious_ECM_fit'

Parsimonious_ECM_diagnostics_test

Return the diagnostic test for 'Parsimonious_ECM_fit'.The diagnostic tests items are the Breusch-Godfrey test for higher-order serial correlation (BG_SC_lm_test). The Engle (1982) test for conditional heteroscedasticity (LM_ARCH_test). The test for non-normality is that of Jarque and Bera (1980). The RESET null hypothesis adopted implies - including the 2nd - degree terms improve the fit (over the model specified). Ljung and Box (1978) tests for autocorrelation in the residuals

cointegration

Return the F statistic, the upper and lower critical values for PSS (2001) bounds test

Longrun_relation

The estimated longrun relation from the error correction model

Arguments

x

data frame

dep_var

A character vector. The dependent variable.

expl_var

Character vector. List of explanatory variable(s)

p_order

Integer. Maximum number of lags for 'dep_var'

q_order

Integer. Maximum number of lags for 'expl_var'

gets_pval

Integer value between 0 and 1 needed for the general-to-specific approach. The default is 0.1 (10 percent significance level). The chosen p-value is the criteria for determining non-significant repressors to be eliminated in a backward elimination path. The final parsimonious model is the best fit model based on the Schwarz information criteria

case

Positive integer 1 to 5. Default is 3

F_HC

Logical (default is FALSE). If TRUE, Heteroscedasticity-Consistent Covariance Matrix Estimation is applied to the model before when estimating F statistic

graph_save

Logical. If TRUE, display stability plot. Default is FALSE

order_l

Integer. order for the serial correlation, and heteroscedasticity test

References

Engle, R. F. (1982). Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflations. Econometrica (50) 987 - 1007

Ljung GM, Box GEP (1978). On a Measure of Lack of Fit in Time Series Models. Biometrika, 65(2), 297 - 303. https://doi.org/10.2307/2335207

Jarque C, Bera A (1980). Efficient Tests for Normality, Homoskedasticity, and Serial Independence. Economics Letters, 6(3), 255 - 259. https://doi.org/10.1016/0165-1765 (80) 90024-5

Pesaran, M. H., Shin, Y., & Smith, R. J. (2001). Bounds testing approaches to the analysis of level relationships Journal of applied econometrics, 16(3), 289-326

See Also

gets_nardl_uecm ardl_uecm

Examples

  data(expectation)
  out <- gets_ardl_uecm(x = expectation,
                        dep_var = c('nq_inf_exp'), 
                        expl_var = c('food_inf','nethawkish'),
                        p_order = c(4), 
                        q_order = c(5,7),
                        gets_pval = 0.1, 
                        case = 4, 
                        graph_save = FALSE,
                        F_HC = FALSE,
                        order_l = 7)
  out