Wavelet estimation of fractional cointegration vector for multivariate time series

Document Type : Research Paper

Authors

Department of Statistics, College of Administration and Economics, University of Baghdad, Iraq

Abstract

Cointegration analysis is one of the most active areas in the econometrics and time series where different methods are introduced for identifying and estimating cointegration vectors in fractionally integrated time series.  In this paper, we estimate the fractional cointegration vector by fully modified narrow band least squares method (FMNBLS) and a proposed method, depending on a linear regression model and wavelet theory, and assuming the errors of the model following  ARFIMA model, also estimate the fractional parameter (long memory parameter) for each variable depending on the Wavelet Whittle method.  These methods were applied on simulated multivariate data for functional magnetic resonance imaging (fMRI) using the R program and programming the proposed method by MATLAB program.

Keywords

[1] S. Achard and I. Gannaz, Wavelet-based and Fourier-based multivariate Whittle estimation multi-wave, J. Statist. Software 89(6) (2018).
[2] S. Achard and I. Gannaz, Multivariate wavelet Whittle estimation in long-range dependence, J. Time Ser. Anal. 37(4) (2016).
[3] M. Demirel and G. Unal, Applying multivariate fractionally integrated volatility analysis on emerging market bond portfolios, Financ. Innov. 6(1) (2020).
[4] M.J. Fadili and E.T. Bullmore, Wavelet-generalized least squares a new BLU estimator of linear regression models with 1 on f errors, NeuroImage 15 (2002) 217–232.
[5] F Sowell, Maximum likelihood estimation of fractionally integrated time series model, Economet. 1990.
[6] B. Jan and T. Norma, Estimation of the long-memory parameter, based on a multivariate central limit theorem, J. Time Ser. Anal. 15(3) (1992).
[7] S. Kechagias and V. Pipiras, Identification, estimation and applications of a bivariate long-range dependent time series model with general phase, preprint, 2018.
[8] K. Liu, Y. Chen and Xi Zhang, An evaluation of ARFIMA (autoregressive fractional integral moving average) programs, Axioms 6(4) (2017).
[9] D. Marinucci, P.M. Robinson, Semiparametric fractional cointegration analysis, J. Economet. 105 (2001) 225–247.
[10] K. Nanamiya, Modelling for the wavelet coefficients of ARFIMA processes, J. Time Ser. Anal. 35 (2014) 341–356.
[11] M.Q. Nielsen, Nonparametric cointegration analysis of fractional systems with unknown integration orders, J. Economet. 155 (2) (2010) 170–187.
[12] M.Q. Nielsen and P. Frederiksen, Fully modified narrow-band least squares estimation of fractional cointegration, Econom. J. 14 (2011) 77–120.
[13] J. Pai and N. Ravishanker, Fast approximate likelihood evaluation for stable VARFIMA processes, Statist. Probab. Lett. 103 (2015) 160–168.
[14] J. Pai and N. Ravishanker, A multivariate preconditioned conjugate gradient approach for maximum likelihood estimation in vector long memory processes, Statist. Probab. Lett. 79 (2009) 1282–1289.
[15] P.M. Robinson, Long-Memory Time Series, Time Series with Long Memory, Oxford University Press, Oxford, 2018.
[16] M.J. Rosa, J. Kilner, F. Blankenburg, O. Josephs and W. Penny, Estimating the transfer function from neuronal activity to BOLD using simultaneous EEG-fMRI, NeuroImage 49(2) (2010) 1496–1509.
[17] P. Sibbertsen, C. Leschinski and M. Busch, A multivariate test against spurious long memory, J. Econom. 203(1) (2018) 33–49.
[18] W. Tsay, Using difference-based methods for inference in regression with fractionally integrated processes, J. Time Ser. Anal. 28(6) (2007).
[19] M. Welvaert, J. Durnez, B. Moerkerke, G. Verdoolaege and Y. Rosseel, R package for generating fMRI data, J. Statist. Software 44 (10) 2011.
Volume 13, Issue 1
March 2022
Pages 3683-3695
  • Receive Date: 07 November 2021
  • Revise Date: 20 December 2021
  • Accept Date: 01 January 2022