Abstract
Let (X i,Y i),i=1, 2,..., be i.i.d. vector valued random variables with unknown common marginal distribution functionsF(x) andG(x). One model of incomplete observations studied in the literature is the truncated model, where bothX i andY i are observed ifX i ≥Y i, and nothing can be observed otherwise. From this kind of observations, if any, we describe the modified nonparametric maximum likelihood estimators ofF(x). The law of the iterated logarithm for the uniform covergence is proved.
Similar content being viewed by others
References
Billingsley, P., Probability and Measure, Wiley, New York, 1986.
Lai, T.L. and Ying, Z., Estimating a Distribution Function with Truncated and Censored Data, Technical Report, 5. Stanford University, 1988.
Lynden-Bell, D.A., A Method of Allowing for Lnown Observational Selection in Small Samples Applied to 3CR Quasars,Mon Nat. Roy. Astr. Soc.,155 (1971), 95–118.
Serfling, R.G., Approximation Theorems of Mathematical Statistics, Wiley, New York, 1980.
Stout, W.F., Almost Sure Convergence, Academic Press, New York, 1974.
Wang, M.C., Jewell, N.P. and Tsai, W.Y., Asymptotic Properties of the Product Limit Estimate under Random Truncation,Ann. Statist.,14 (1986), 1597–1605.
Woodroofe, M., Estimating a distribution with Truncated Data,Ann. Statist.,13 (1985), 163–177.
Author information
Authors and Affiliations
Additional information
This work is supported by the National Natural Science Foundation of China.
Rights and permissions
About this article
Cite this article
He, S. Estimating a distribution function with truncated data. Acta Mathematicae Applicatae Sinica 10, 12–33 (1994). https://doi.org/10.1007/BF02006256
Received:
Issue Date:
DOI: https://doi.org/10.1007/BF02006256