Communications for Statistical Applications and Methods

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

The Comparison of Imputation Methods in Space Time Series Data with Missing Values

공간시계열모형의 결측치 추정방법 비교

  • Lee, Sung-Duck (Department of Information and Statistics, Chungbuk National University) ;
  • Kim, Duck-Ki (Department of Information and Statistics, Chungbuk National University)
  • Received : 20100100
  • Accepted : 20100300
  • Published : 2010.03.31
Download PDF
⟨ Previous Next ⟩

Abstract

Missing values in time series can be treated as unknown parameters and estimated by maximum likelihood or as random variables and predicted by the conditional expectation of the unknown values given the data. The purpose of this study is to impute missing values which are regarded as the maximum likelihood estimator and random variable in incomplete data and to compare with two methods using ARMA and STAR model. For illustration, the Mumps data reported from the national capital region monthly over the years 2001~2009 are used, and estimate precision of missing values and forecast precision of future data are compared with two methods.

시계열의 결측값은 미지의 모수 또는 확률변수로 취급할 수 있으며 이에 따른 최대가능도방법과 확률변수방법에 의해 결측치를 추정할수 있으며 또한 주어진 자료 하에서 미지의 값에 대한 조건부기대치로 예측할수 있다. 이 연구의 주된 목적은 불완전한 자료에 대해 기존에는 ARMA모형만을 고려하였는데 이를 확장하여 공간시계열모형인 STAR모형에 적용하여 두 가지 추정방법을 이용해 결측값의 추정 정밀도를 비교하는데 있다. 사례분석을 위해 한국질병관리본부에서 전산보고 하고 있는 전염병 자료 중에서 2001~2009년 동안의 월별 Mumps 자료를 이용하여 두 가지 추정방법의 추정 정밀도와 예측정확도를 비교하였다.

Keywords

Acknowledgement

Supported by : 충북대학교

References

  1. 손건태, 백지선 (1997). 공간자기회귀모형의 식별, <응용통계학회>, 10, 121-136.
  2. 이성덕, 김덕기 (2009). 시계열자료에서 결측치 추정방법의 비교, <한국통계학회>, 16, 723-730. https://doi.org/10.5351/CKSS.2009.16.4.723
  3. Alcroft, D. J. and Glasbey, C. A. (2005). STARMA process applied to Solar Radiation, Biomathmatics and Statistics Scotland, 1-24.
  4. Bayarri, M. J., DeGroot, M. H. and Kadane, J. B. (1986). What is the likelihood function?, In: Statistical Decision Theory and Related Topics IV, Volume 1., Sparinger-Verlag, New York.
  5. Billard, L. and Dai, Y. (1998). A space-time bilinear model and its identification, Journal of Time Series Analysis, 19, 657-679. https://doi.org/10.1111/1467-9892.00115
  6. Billard, L. and Dai, Y. (2003). Maximum likelihood estimation in space time bilinear model, Journal of Time Serise Analysis, 24, 25-44. https://doi.org/10.1111/1467-9892.00291
  7. Brubacher, S. R. and Wilson, T. (1976). Interpolating time series with application to the estimation of holiday effects on electricity demand, Applied statistics, 25, 107-116. https://doi.org/10.2307/2346678
  8. Dunsmuir, W. and Robinson, P. M. (1981). Estimation of time series models in the presence of missing data, Journal of the American Statistical Association, 76, 560-568. https://doi.org/10.1080/01621459.1981.10477687
  9. Horton, N. J. and Kleinman, K. P. (2007). A comparison of missing data methods and software to fit incomplete data regressing models, The American Statistication, 61, 79-90. https://doi.org/10.1198/000313007X172556
  10. Kim, J. (2005). Parameter estimation in stochastic volatility models with missing data using particle methods and the EM algorithm, PhD thesis, University of pittsburgh.
  11. Pena, D. (1987). Measuring the importance of outliers in ARIMA models, In New Perspectives in Theoretical and Applied Statistics, M. L. Puri (ed). Wiley, New York, 109-118.
  12. Pena, D. and Tiao, G. C. (1991). A note on likelihood estimation of missing values in time series, The American Statisticaian, 45, 212-213. https://doi.org/10.2307/2684292
  13. Pfeifer, P. E. and Deutsch, S. J. (1980). Identification and interpretation of first order space-time ARMA models, Technometrics, 22, 397-408. https://doi.org/10.1080/00401706.1980.10486172
  14. Rubin, D. B. (1994). Missing data, imputation, and the bootstrap: Comment, Journal of the American Statistical Association, 89, 475-478. https://doi.org/10.2307/2290847

Cited by

  1. Robustizing Kalman filters with the M-estimating functions vol.25, pp.1, 2018, https://doi.org/10.29220/CSAM.2018.25.1.099
  2. A Missing Value Replacement Method for Agricultural Meteorological Data Using Bayesian Spatio–Temporal Model vol.27, pp.7, 2018, https://doi.org/10.5322/JESI.2018.27.7.499