About The Project

This repository provides C++ codes used to conduct Monte Carlo simulations and empirical applications in the paper: A Distance-Based Test of Independence Between two Multivariate Time Series

Pre-requisites

Build

On an Ubuntu machine, a binary executable can be built from a C++ main file by running following shell script from the terminal:

g++ -Wno-deprecated -O3 -Wfloat-equal -Wfatal-errors -m64 -std=gnu++17 -fopenmp -lshogun -lspdlog -lboost_thread -Wunknown-pragmas -Wall -Waggressive-loop-optimizations -mavx2 -march=native -I/<path-to-the-folder-containing-source-codes>/ -I/usr/include -I/usr/lib/gcc/x86_64-linux-gnu/4.9.3/include -I/<path-to-plotcpp-library> -I/<path-to-gsl-2.7>/include -I/<path-to-shogun-library>/include -I/usr/include/eigen3 -I/usr/local/include -c/<path-to-the-folder-containing-source-codes>/<main file with extension *.cpp> -o .objs/main.o

g++ -L/<path-to-gsl-2.7>/lib -L/<path-to-shogun-library>/lib -L/usr/lib/x86_64-linux-gnu -L/usr/lib -o <name-of-the-binary-to-be-built> .objs/main.o  -fopenmp -O3 -m64 -lshogun -lspdlog  -lgsl -lgslcblas -lm

List of C++ main files

C++ main file	Description
`main_VAR_CC_MGARCH_MGARCH_data.cpp`	to simulate the proposed test with the true DGP CC_MGARCH(1) to generate and fit data
`main_VAR_CC_MGARCH_VAR_data.cpp`	to simulate the proposed test with the true DGP VAR(1) to generate and fit data
`main_VAR1_updated.cpp`	to simulate the proposed test using VAR(1) to generate data and Random Forest (RF) to fit data
`main_CC_MGARCH.cpp`	to simulate the proposed test using CC-MGARCH(1, 1) to generate data and RF to fit data
`main_others_VAR1.cpp`	to simulate the other tests using VAR(1) to generate and fit data
`main_others_CC_MGARCH.cpp`	to simulate the other tests using CC-MGARCH(1, 1) to generate and fit data
`main_others_VAR1_misspec.cpp`	to simulate the other tests using VAR(1) to generate data and CC-MGARCH(1, 1) to fit data
`main_others_CC_MGARCH_misspec.cpp`	to simulate the other tests using CC-MGARCH(1, 1) to generate data and VAR(1) to fit data
`main_bootstrp_using_MGARCH_Stock_app.cpp`	to implement the proposed bootstrap test using RF in the empirical application (stocks vs. bonds)
`main_timing_hsic_dcorr_tests.cpp`	to time the proposed test and the HSIC-based test
`main_Beta_bivariate.cpp`	to simulate the proposed test for the bivariate case using the Beta errors
`main_Beta_univariate.cpp`	to simulate the proposed test for the univariate case using the Beta errors
`main_Exp_bivariate.cpp`	to simulate the proposed test for the bivariate case using the exponential errors
`main_Exp_univariate.cpp`	to simulate the proposed test for the univariate case using the exponential errors
`main_MN_bivariate.cpp`	to simulate the proposed test for the bivariate case using the mixtures of standard normal errors
`main_MN_univariate.cpp`	to simulate the proposed test for the univariate case using the mixtures of standard normal errors
`main_SN_bivariate.cpp`	to simulate the proposed test for the bivariate case using the skew-normal errors
`main_SN_univariate.cpp`	to simulate the proposed test for the univariate case using the skew-normal errors
`main_Beta_others.cpp`	to simulate all the other tests using the Beta errors
`main_Exp_others.cpp`	to simulate all the other tests using the exponential errors
`main_MN_others.cpp`	to simulate all the other tests using the mixtures of standard normal errors
`main_SN_others.cpp`	to simulate all the other tests using the skew-normal errors

Note: You may obtain some numbers that are slightly different from the numbers reported in the paper, but these differences will not change the main findings reported in the Monte-Carlo study section. The reason is that we use a GSL random number generator algorithm to generate random samples from known probability distributions, and this algorithm employs hardware configurations and interrupts so that random numbers are actually hardware-dependent. Also, to ensure that the generated random samples are distinct and have maximum randomness, we use random seeds (gsl_rng_get) in Monte-Carlo loops.

License

Distributed under the MIT License. See LICENSE.txt for more information.

Contact

Ba Chu - ba.chu@carleton.ca

Project Link: https://github.com/wave1122/DcorrTest1

Name		Name	Last commit message	Last commit date
Latest commit History 5 Commits
ShogunML		ShogunML
data/Bond		data/Bond
.gitattributes		.gitattributes
.gitignore		.gitignore
LICENSE.txt		LICENSE.txt
ML_dcorr.h		ML_dcorr.h
ML_dcorr_power.h		ML_dcorr_power.h
ML_dcorr_power.h.save-failed		ML_dcorr_power.h.save-failed
ML_reg_dcorr.h		ML_reg_dcorr.h
ML_reg_dcorr_ver1.h		ML_reg_dcorr_ver1.h
Makefile.win		Makefile.win
Matrix.h		Matrix.h
NumMeth.h		NumMeth.h
README.md		README.md
RF.h		RF.h
The_rmgarch_models.pdf		The_rmgarch_models.pdf
VAR_CC_MGARCH_dcorr_power.h		VAR_CC_MGARCH_dcorr_power.h
VAR_CC_MGARCH_reg_dcorr.h		VAR_CC_MGARCH_reg_dcorr.h
VAR_CC_MGARCH_reg_dcorr.h.save-failed		VAR_CC_MGARCH_reg_dcorr.h.save-failed
VAR_gLasso.h		VAR_gLasso.h
asserts.h		asserts.h
cross_validation_timeseries_split.dat		cross_validation_timeseries_split.dat
csvreader.h		csvreader.h
dcorr		dcorr
dcorr1		dcorr1
dep_tests.h		dep_tests.h
dep_tests1.h		dep_tests1.h
dgp.h		dgp.h
dist_corr.h		dist_corr.h
empirical_app		empirical_app
exogeneity.cbp		exogeneity.cbp
exogeneity.depend		exogeneity.depend
exogeneity.dev		exogeneity.dev
exogeneity.layout		exogeneity.layout
exogeneity.mk		exogeneity.mk
exogeneity.project		exogeneity.project
exogeneity.workspace		exogeneity.workspace
filein.h		filein.h
gaussian_reg.h		gaussian_reg.h
gtrendsR.pdf		gtrendsR.pdf
histogram.R		histogram.R
hsic_test.h		hsic_test.h
kernel.h		kernel.h
lconv_intl_refcount		lconv_intl_refcount
list.log		list.log
local.h		local.h
main.cpp		main.cpp
main.cpp.save-failed		main.cpp.save-failed
main_Beta_bivariate.cpp		main_Beta_bivariate.cpp
main_Beta_others.cpp		main_Beta_others.cpp
main_Beta_univariate.cpp		main_Beta_univariate.cpp
main_CC_MGARCH.cpp		main_CC_MGARCH.cpp
main_CC_MGARCH_bootstrap.cpp		main_CC_MGARCH_bootstrap.cpp
main_ChiSq_bivariate.cpp		main_ChiSq_bivariate.cpp
main_ChiSq_others.cpp		main_ChiSq_others.cpp
main_ChiSq_univariate.cpp		main_ChiSq_univariate.cpp
main_Exp_bivariate.cpp		main_Exp_bivariate.cpp
main_Exp_others.cpp		main_Exp_others.cpp
main_Exp_univariate.cpp		main_Exp_univariate.cpp
main_MN_bivariate.cpp		main_MN_bivariate.cpp
main_MN_others.cpp		main_MN_others.cpp
main_MN_univariate.cpp		main_MN_univariate.cpp
main_SN_bivariate.cpp		main_SN_bivariate.cpp
main_SN_multivariate.cpp		main_SN_multivariate.cpp
main_SN_others.cpp		main_SN_others.cpp
main_SN_univariate.cpp		main_SN_univariate.cpp
main_VAR1.cpp		main_VAR1.cpp
main_VAR1_bootstrap.cpp		main_VAR1_bootstrap.cpp
main_VAR1_updated.cpp		main_VAR1_updated.cpp
main_VAR_CC_MGARCH_MGARCH_data.cpp		main_VAR_CC_MGARCH_MGARCH_data.cpp
main_VAR_CC_MGARCH_VAR_data.cpp		main_VAR_CC_MGARCH_VAR_data.cpp
main_bootstrp_Beta_bivariate.cpp		main_bootstrp_Beta_bivariate.cpp
main_bootstrp_Exp_bivariate.cpp		main_bootstrp_Exp_bivariate.cpp
main_bootstrp_MN_bivariate.cpp		main_bootstrp_MN_bivariate.cpp
main_bootstrp_SN_bivariate.cpp		main_bootstrp_SN_bivariate.cpp
main_bootstrp_using_MGARCH.cpp		main_bootstrp_using_MGARCH.cpp
main_bootstrp_using_MGARCH_Covid_app.cpp		main_bootstrp_using_MGARCH_Covid_app.cpp
main_bootstrp_using_MGARCH_GDP_app.cpp		main_bootstrp_using_MGARCH_GDP_app.cpp
main_bootstrp_using_MGARCH_GT_app.cpp		main_bootstrp_using_MGARCH_GT_app.cpp
main_bootstrp_using_MGARCH_Oil_app.cpp		main_bootstrp_using_MGARCH_Oil_app.cpp
main_bootstrp_using_MGARCH_Stock_app.cpp		main_bootstrp_using_MGARCH_Stock_app.cpp
main_gaussian.cpp		main_gaussian.cpp
main_gaussian_bivariate.cpp		main_gaussian_bivariate.cpp
main_gaussian_mixedAR.cpp		main_gaussian_mixedAR.cpp
main_gaussian_test.cpp		main_gaussian_test.cpp
main_nongaussian_bivariate.cpp		main_nongaussian_bivariate.cpp
main_others_CC_MGARCH.cpp		main_others_CC_MGARCH.cpp
main_others_CC_MGARCH_misspec.cpp		main_others_CC_MGARCH_misspec.cpp
main_others_VAR1.cpp		main_others_VAR1.cpp
main_others_VAR1_misspec.cpp		main_others_VAR1_misspec.cpp
main_studentt.cpp		main_studentt.cpp
main_test_El_Himdi_Roy.cpp		main_test_El_Himdi_Roy.cpp
main_test_using_gLASSO.cpp		main_test_using_gLASSO.cpp
main_timing_hsic_dcorr_tests.cpp		main_timing_hsic_dcorr_tests.cpp
matrix_ops2.h		matrix_ops2.h
matrix_ops2.h.bk		matrix_ops2.h.bk
mgarch.h		mgarch.h
mgarch_max.h		mgarch_max.h
mymodel.xml		mymodel.xml
ndgp.h		ndgp.h

License

wave1122/DcorrTest1

Folders and files

Latest commit

History