On estimation of parameters in the case of discontinuous densities

A. A. Borovkov

Teoriâ veroâtnostej i ee primeneniâ, Tome 63 (2018) no. 2, pp. 211-239

Résumé

This paper is concerned with the problem of construction of estimators of parameters in the case when the density $f_\theta(x)$ of the distribution $\mathbf{P}_\theta$ of a sample $\mathrm X$ of size $n$ has at least one point of discontinuity $x(\theta)$, $x'(\theta)\neq 0$. It is assumed that either (a) from a priori considerations one can specify a localization of the parameter $\theta$ (or points of discontinuity) satisfying easily verifiable conditions, or (b) there exists a consistent estimator $\widetilde{\theta}$ of the parameter $\theta$ (possibly constructed from the same sample $\mathrm{X}$), which also provides some localization. Then a simple rule is used to construct, from the segment of the empirical distribution function defined by the localization, a family of estimators $\theta^*_{g}$ that depends on the parameter $g$ such that (1) for sufficiently large $n$, the probabilities $\mathbf{P}(\theta^*_{g}-\theta>v/n)$ and $\mathbf{P}(\theta^*_{g}-\theta<-v/n)$ can be explicitly estimated by a $v$-exponential bound; (2) in case (b) under suitable conditions (see conditions I–IV in Chap. 5 of [I. A. Ibragimov and R. Z. Has'minskiĭ, Statistical Estimation. Asymptotic Theory, Springer, New York, 1981], where maximum likelihood estimators were studied), a value of $g$ can be given such that the estimator $\theta^*_{g}$ is asymptotically equivalent to the maximum likelihood estimator $\widehat{\theta}$; i.e., $\mathbf{P}_\theta(n(\theta^*_{g}-\theta)>v)\sim \mathbf{P}_\theta(n(\widehat{\theta}-\theta)>v)$ for any $v$ and $n\to\infty$; (3) the value of $g$ can be chosen so that the inequality $\mathbf{E}_\theta(\theta^*_{g}-\theta)^2< \mathbf{E}_\theta(\widehat{\theta}-\theta)^2$ is possible for sufficiently large $n$. Effectively no smoothness conditions are imposed on $f_\theta(x)$. With an available “auxiliary” consistent estimator $\widetilde{\theta}$, simple rules are suggested for finding estimators $\theta^*_g$ which are asymptotically equivalent to $\widehat{\theta}$. The limiting distribution of $n(\theta^*_g-\theta)$ as $n\to\infty$ is studied.

Détail

Keywords: estimators of parameters, maximum likelihood estimator, distribution with discontinuous density, change-point problem, infinitely divisible factorization.

Bibliographie

[1] I. A. Ibragimov, R. Z. Has'minskiĭ, Statistical estimation. Asymptotic theory, Appl. Math., 16, Springer-Verlag, New York–Berlin, 1981, vii+403 pp. | MR | MR | Zbl | Zbl

[2] A. A. Borovkov, Mathematical statistics, Gordon Breach Science Publishers, Amsterdam, 1998, xxii+570 pp. | MR | Zbl

[3] M. S. Ermakov, “Asymptotic behavior of parameter estimators of a multivariate discontinuous density”, J. Soviet Math., 34:1 (1986), 1413–1427 | DOI | MR | Zbl

[4] V. E. Mosyagin, “An estimate of the rate of convergence of the distribution of the maximum probability process in the nonregular case”, Sib. Math. J., 32:4 (1991), 616–622 | DOI | MR | Zbl

[5] V. E. Mosyagin, “Asymptotic representation of the process of the likelihood ratio in the case of a discontinuous density”, Sib. Math. J., 35:2 (1994), 375–382 | DOI | MR | Zbl

[6] V. E. Mosyagin, “Estimation of the convergence rate for the distributions of normalized maximum likelihood estimators in the case of a discontinuous density”, Sib. Math. J., 37:4 (1996), 788–796 | DOI | MR | Zbl

[7] I. S. Borisov, D. V. Mironov, “An asymptotic representation for the likelihood ratio for multidimensional samples with discontinuous desities”, Theory Probab. Appl., 45:2 (2001), 289–300 | DOI | DOI | MR | Zbl

[8] I. S. Borisov, D. V. Mironov, “An asymptotic representation of the likelihood ratio for irregular families of distributions in the multivariate case”, Sib. Math. J., 42:2 (2001), 232–244 | DOI | MR | Zbl

[9] A. A. Borovkov, Stochastic processes in queueing theory, Appl. Math., 4, Springer-Verlag, New York–Berlin, 1976, xi+280 pp. | MR | MR | Zbl | Zbl

[10] E. A. Pecherskii, B. A. Rogozin, “On joint distributions of random variables associated with fluctuations of a process with independent increments”, Theory Probab. Appl., 14:3 (1969), 410–423 | DOI | MR | Zbl

[11] N. S. Bratiichuk, D. V. Gusak, Granichnye zadachi dlya protsessov s nezavisimymi prirascheniyami, Naukova dumka, Kiev, 1990, 264 pp. | MR | Zbl

[12] A. V. Skorokhod, Random processes with independent increments, Math. Appl. (Soviet Ser.), 47, Kluwer Acad. Publ., Dordrecht, 1991, xii+279 pp. | MR | MR | Zbl | Zbl

[13] V. E. Mosyagin, N. A. Shvemler, “Raspredelenie momenta maksimuma raznosti dvukh puassonovskikh protsessov s otritsatelnym lineinym snosom”, Sib. elektron. matem. izv., 13 (2016), 1229–1248 | DOI | MR | Zbl

[14] A. A. Borovkov, “New limit theorems in boundary problems for sums of independent terms”, Select. Transl. Math. Stat. Probab., 5, Amer. Math. Soc., Providence, RI, 1965, 315–372 ; “II”, 5, no. 4, 377–392 | MR | MR | Zbl | Zbl | MR | Zbl

[15] A. A. Borovkov, “Limit theorems on the distributions of maxima of sums of bounded lattice random variables. I”, Theory Probab. Appl., 5:2 (1960), 125–155 ; “II”, 5:4, 341–355 | DOI | DOI | MR | Zbl | Zbl