Shrinkage Estimation of Linear Regression Models with ARIMA Errors and Applications to Canadian Crime Rates Data
DOI:
https://doi.org/10.6092/issn.1973-2201/19747Keywords:
Linear regression model, ARIMA, Monte Carlo simulation, Shrinkage estimators, Asymptotic biases and risksAbstract
Shrinkage methods for estimating the parameters of a regression model with autoregressive integrated moving average (ARIMA) errors are presented when some of the regression parameters are restricted to a subspace. The estimates are defined by the maximization of the likelihood function with and without restriction which makes the unrestricted and restricted estimators, respectively. The shrinkage estimators combine these two estimators optimally. To demonstrate the optimality of these estimators, we rely on metrics like asymptotic distributional bias (ADB) or asymptotic distributional risk (ADR), which aim to minimize these two quantities. We show that the relative efficiency of the shrinkage estimator is superior to that of the unrestricted estimator when the dimension of shrinkage exceeds two. Our large sample theory and simulation study demonstrate that shrinkage estimators dominate the unrestricted estimator in the entire parameter space. An empirical example of Canadian crime rates data is also provided.
References
A. M. BIANCO, G. C. BEN, M. G. AND. MARTINEZ, V. YOHAI (2001). Outlier detection in regression models with ARIMA errors using robust estimates. Journal of Forecasting, 20, pp. 565–579.
K.-S. CHAN, K. CHEN (2011). Subset ARMA selection via the adaptive LASSO. Statistics and its Interface, 4, no. 2, pp. 197–205.
S. CHEN, X. WANG, J. ZHAO, Y. ZHANG, X. KAN (2022). Application of the ARIMA model in forecasting the incidence of tuberculosis in anhui during COVID-19 pandemic from 2021 to 2022. Infection and Drug Resistance, 15, pp. 3503–3512.
R. A. DAVIS, W. T. M. DUNSMUIR (1997). Least absolute deviation estimation for regression with ARMA errors. Journal of Theoretical Probability, 10, no. 2, pp. 481–497.
J. FOX, S.WEISBERG (2019). An R Companion to Applied Regression. Sage publications, Thousand Oaks CA.
M. FURNO (1996). The information matrix test in the linear regression with ARMA errors. Journal of the Italian Statistical Society, 5, pp. 369–385.
E. GEORGE, G. JENKINS, G. REINSEL (2008). Time Series Analysis: Forecasting and Control. JohnWiley and Sons, New Jersey.
A. K. GUPTA, A. M. E. SALEH, P. SEN (1989). Improved estimation in a contingency table: independence structure. Journal of the American Statistical Association, 84, no. 406, pp. 525–532.
W. HÄRDLE, J. HOROWITZ, J. P. KREISS (2003). Bootstrap methods for time series. International Statistical Review, 71, no. 2, pp. 435–459.
S.HOSSAIN, M. GHAHRAMANI (2016). Shrinkage estimation of linear regression models with GARCH errors. Journal of Statistical Theory and Applications, 15, no. 4, pp. 405–423.
S. HOSSAIN, L. A. LAC (2018). Non-penalty shrinkage estimation of random effect models for longitudinal data with AR(1) errors. Journal of Statistical Computation and Simulation, 88, no. 16, pp. 3230–3247.
S. JOHANSEN, M. RIANI, A. ATKINSON (2012). The selection of ARIMA models with or without regressors. University of Copenhagen, Department of Economics Discussion Paper, 12-17, pp. 1–32.
M. C. OTTO, W. R. BELL, J. P. BURMAN (1987). An Iterative GLS Approach to Maximum Likelihood Estimation of Regression Models with ARIMA Errors. Research Report No. 87/34, Statistical Research Division, US Bureau of the Census,Washington, D.C.
A. PANKRATZ (1991). Forecasting With Dynamic Regression Models. John Wiley and Sons, New York.
M. PAOLELLA (2019). Linear Models and Time Series Analysis: Regression, ANOVA, ARMA and GARCH. John Wiley and Sons, New York.
J. H. ROB, A. GEORGE (2021). Forecasting: Principles and Practice. OTexts, Melbourne, Australia.
P. K. SEN, A. K. M. EHSANES SALEH (1987). On preliminary test and shrinkage Mestimation in linear models. The Annals of Statistics, 15, no. 4, pp. 1580–1592.
T. THOMSON, S. HOSSAIN, M. GHAHRAMANI (2015). Application of shrinkage estimation in linear regression models with autoregressive errors. Statistical Computation and Simulation, 85, no. 16, pp. 3335–3351.
T. THOMSON, S. HOSSAIN, M. GHAHRAMANI (2016). Efficient estimation for time series following generalized linear models. Australian&New Zealand Journal of Statistics, 58, no. 4, pp. 493–513.
H.WANG, G. LI, C. TSAI (2007). Regression coefficient and autoregressive order shrinkage and selection via the LASSO. Journal of the Royal Statistical Society. Series B (Statistical Methodology), 69, no. 1, pp. 63–78.
R. B. WILLIAM (2011). REGCMPNT–A fortran program for regression models with ARIMA component errors. Journal of Statistical Software, 41, no. 7, pp. 1–23.
M. A. WINCEK, G. C. REINSEL (1986). An exact maximum likelihood estimation procedure for regression ARMA time series models with possibly nonconsecutive data. Journal of the Royal Statistical Society. Series B (Statistical Methodology), 48, no. 3, pp. 303–313.
R. WU, Q. WANG (2012). Shrinkage estimation for linear regression with ARMA errors. Journal of Statistical Planning and Inference, 142, no. 7, pp. 2136–2148.
M. XU, Z. QIN (2021). A novel hybrid ARIMA and regression tree model for the interval-valued time series. Journal of Statistical Computation and Simulation, 91, no. 5, pp. 1000–1015.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2024 Statistica

This work is licensed under a Creative Commons Attribution 4.0 International License.