A Suggested Estimator for AR(1) Model with Missing Observations
Keywords:
AR(1), missing observations, stationary, unconditional maximum likelihoodAbstract
The paper considers estimation of stationary first-order autoregressive model AR(1) with missing observations. The maximum likelihood method is used to estimate the autoregressive parameter for AR(1) with missing observations. The efficiency of the estimation is affected by treating the initial value required to compute the first value of residuals. The conventional methods treat the initial value as fixed. Therefore, we present new method to estimate AR(1) with missing observations based on treating the initial value as random. The likelihood function is uniquely maximized and a new closed-form estimator for AR(1) in case of missing observations is developed. Monte Carlo simulation studies and a real data analysis showed that the bias and efficiency of the new estimators are more reliable than the conventional estimators. Moreover, the proposed method provides better estimates of missing values than the existing methods.
References
Abdelwahab MM. On Parameter Estimation of Time Series Models with Missing Observations. PhD [thesis]. Egypt: Cairo University; 2016.
Abdelwahab MM, Issa MK. Forms of the moments of AR(P) model with missing observations. Egyptian Stat J. 2019; 63(1): 39-44.
Box GEP, Jenkins GM. Time series analysis: Forecasting and control. San Francisco: Holden-Day; 1976.
Ding J, Han L, Chen X. Time series AR modeling with missing observations based on the polynomial transformation. Math Comput Model. 2010; 51(5-6): 527-536.
Dunsmuir W, Robinson PM. Estimation of time series models in the presence of missing data. J Am Stat Assoc. 1981; 76(375): 560-568.
El-Sayed SM, El-Sheikh AA, Abdelwahab MM. Estmation of AR(1) models with missing values. Adv Appl Stat. 2016; 49(6): 485-492.
Enany, MA, Issa MK, Gad AA. Maximum likelihood estimation of unconditional AR(1) model with missing observations. Working paper. Egypt: Zagazig University; 2020.
Hamaz A, Ibazizen M. Comparison of two estimation methods of missing values using Pitman-closeness criterion. Commun Stat Theory Methods. 2009; 38(13): 2210-2213.
Hasza DP. A note on maximum likelihood estimation for the first-order autoregressive process. Commun Stat Theory Methods. 1980; 9(13): 1411-1415.
Nicholson AJ. Population oscillations caused by competition for food. Nature. 1950; 165(4195): 476-477.
Parzen E. On spectral analysis with missing observations and amplitude modulation. Sankhyā, Series A. 1963; 25: 383-392.
Pourahmadi M. Estimation and interpolation of missing values of a stationary time series. J Time Anal. 1989; 10(2): 149-169.
Shumway RH, Stoffer DS. Time series analysis and its applications: with R examples. New York: Springer; 2017.
Singh, S. Advanced Sampling Theory with Applications. Netherlands: Kluwer Academics Press; 2003.
Takeuchi K. A comment on “Recent Development of Economic Data Analysis” at the 63rd Annual Meeting of Japan Statistical Society; 1995.
Wei WWS. Time series analysis univariate and multivariate methods. New York: Addison Wesley; 2006.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
