Estimasi Distribusi Mixing dalam Model Mixture Poisson

N Dwidayati(1),


(1) Jurusan Matematika, FMIPA, Universitas Negeri Semarang, Indonesia

Abstract

Model mixture telah dikenalkan sejak tahun 1986 (Farewell, 1986) dengan menganalisis penderita kanker payudara. Model ini diaplikasikan dalam ruang kerja yang berbeda sesuai dengan fleksibilitas pandangan kompleksnya situasi, walaupun generalisasi teori belum sepenuhnya dikembangkan. Studi klinik yang dilakukan difokuskan pada estimasi proporsi pasien yang sembuh dan distribusi  failure time pasien yang tidak sembuh. Pada penelitian ini dikonstruksi estimator konsisten dari distribusi mixing dalam model mixture Poisson melalui inversi Laplace, baik untuk data tak tersensor maunpun tersensor. Berdasar estimasi tersebut ditentukan IMSE dan laju kekonvergenan yang berkorespondensi dengan inverse estimator. 

The mixture model has been introduced since 1986 (Farewell, 1986) by analyzing breast cancer patients. This model was applied in different workspaces according to the flexibility of the complex view of the situation, although the generalization of the theory has not been fully developed. Clinical studies undertaken focused on estimating the proportion of patients who recovered and the distribution of patient failure times that did not heal. In this research constructed a consistent estimator of the mixing distribution in the Poisson mixture model through Laplace inversion, both for uncensored and uncensored data. Based on these estimates, the IMSE and the convergence rate correspond to the inverse estimator. 

Keywords

mixing distribution. Poisson mixture model, Laplace inversion, IMSE, convergence

Full Text:

PDF

References

Fan, J. 1991. On The Optimal Rates of Convergence for Nonparametric Deconvolution Problems. Ann.Statist 19: 1257-1272

Farewell, V.T. 1986. Mixture models in survival analysis. Are they worth the risk ? The Canadian Journal of Statistics 14: 257-262.

Feller, W. 1971. An Introduction to Probability Theory and Its Applications,Vol.II. New York: Wiley

Jewel, NP. 1982. Mixtures of Exponential Distribution. Ann Statist. 10:479-484

Klaassen, C.A.J & Mnatsakanov, R.M. 2003. Estimating the Structural Distribution Function of Cell Probabilities. Austrian J.Statist 32: 85-98

Laird, NM. 1978. Nonparametric Maximum Likelihood Estimation of a Mixing Distribution. J.A.S.A. 73: 805-811

Leonard, T. 1970. Some Data-analytic Modifications to Bayes-Stein Estimation. Ann.Inst.Statist. Math Part A.36: 11-21

Lindsay, BG. 1989. Moment Matrics: Applications in Mixture. Ann. Statist 17: 722-740

Loh, WL & Zhang, CH. 1996. Global Properties of Kernel Estimators for Mixing Densities in Discrete Exponential Family Models. Statis. Sinica. 6: 561-578

Meeden, G. 1993. Noninformative Nonparametric Bayesian Estimation on Quantiles. Statistics & Probability Letters 16: 103-109

Pfanzagl, J. 1988. Consistency of Maximum Likelihood for Certain Nonparametric Families, in Particular: Mixture. J. Stat. Plan. Inf. 19: 137-158

Redner, R. 1981. Not on the Consistency of the Maximum Likelihood Estimate for Non Identifiable Distribution. Ann. Stat. 9: 225-228

Simar, L. 1976. Maximum Likelihood Estimation of a Compound Poisson Process. Ann. Statist 4: 1200-1209

Refbacks

  • There are currently no refbacks.




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