Robust Econometric Inference.

ivx: Robust Econometric Inference

Drawing statistical inference on the coefficients of a short- or long-horizon predictive regression with persistent regressors by using the IVX method of Magdalinos and Phillips (2009) and Kostakis, Magdalinos and Stamatogiannis (2015).

Installation

You can install the development version from GitHub with:

# Install release version from CRAN
install.packages("ivx")


# install.packages("devtools")
devtools::install_github("kvasilopoulos/ivx")

Usage

library(ivx)
library(magrittr)

This is a basic example, lets load the data first:

# Monthly data from Kostakis et al (2014)
kms %>%
  names()
#>  [1] "Date" "DE"   "LTY"  "DY"   "DP"   "TBL"  "EP"   "BM"   "INF"  "DFY" 
#> [11] "NTIS" "TMS"  "Ret"

Univariate

And then do the univariate estimation:

ivx(Ret ~ DP, data = kms) %>% 
  summary()
#> 
#> Call:
#> ivx(formula = Ret ~ DP, data = kms, horizon = 1)
#> 
#> Coefficients:
#>    Estimate Wald Ind Pr(> chi)
#> DP 0.006489    2.031     0.154
#> 
#> Joint Wald statistic:  2.031 on 1 DF, p-value 0.1541
#> Multiple R-squared:  0.002844,   Adjusted R-squared:  0.001877

ivx(Ret ~ DP, data = kms, horizon = 4) %>% 
  summary()
#> 
#> Call:
#> ivx(formula = Ret ~ DP, data = kms, horizon = 4)
#> 
#> Coefficients:
#>    Estimate Wald Ind Pr(> chi)
#> DP 0.006931    2.271     0.132
#> 
#> Joint Wald statistic:  2.271 on 1 DF, p-value 0.1318
#> Multiple R-squared:  0.01167,    Adjusted R-squared:  0.01358

Multivariate

And the multivariate estimation, for one or multiple horizons:

ivx(Ret ~ DP + TBL, data = kms) %>% 
  summary()
#> 
#> Call:
#> ivx(formula = Ret ~ DP + TBL, data = kms, horizon = 1)
#> 
#> Coefficients:
#>      Estimate Wald Ind Pr(> chi)
#> DP   0.006145    1.819     0.177
#> TBL -0.080717    1.957     0.162
#> 
#> Joint Wald statistic:  3.644 on 2 DF, p-value 0.1617
#> Multiple R-squared:  0.004968,   Adjusted R-squared:  0.003036

ivx(Ret ~ DP + TBL, data = kms, horizon = 4) %>% 
  summary()
#> 
#> Call:
#> ivx(formula = Ret ~ DP + TBL, data = kms, horizon = 4)
#> 
#> Coefficients:
#>      Estimate Wald Ind Pr(> chi)
#> DP   0.006579    2.045     0.153
#> TBL -0.073549    1.595     0.207
#> 
#> Joint Wald statistic:  3.527 on 2 DF, p-value 0.1715
#> Multiple R-squared:  0.018,  Adjusted R-squared:  0.01895

Yang et al. (2020) IVX-AR methodology

ivx_ar(hpi ~ cpi, data = ylpc) %>% 
  summary()
#> 
#> Call:
#> ivx_ar(formula = hpi ~ cpi, data = ylpc, horizon = 1)
#> 
#> Auto () with AR terms q = 4
#> 
#> Coefficients:
#>       Estimate Wald Ind Pr(> chi)  
#> cpi -0.0001775    4.326    0.0375 *
#> ---
#> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> 
#> Joint Wald statistic:  4.326 on 1 DF, p-value 0.03753
#> Multiple R-squared:  0.02721,    Adjusted R-squared:  0.02142
#> Wald AR statistic: 132.3 on 4 DF, p-value < 2.2e-16

Please note that the ‘ivx’ project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.