We consider the empirical risk function (for iid ’s) under the assumption that f(α,z) is convex with respect to α. Asymptotics of the minimum of is investigated. Tests for linear hypotheses are derived. Our results generalize some of those concerning LAD estimators and related tests.
Statistical inference procedures based on least absolute deviations involve estimates of a matrix which plays the role of a multivariate nuisance parameter. To estimate this matrix, we use kernel smoothing. We show consistency and obtain bounds on the rate of convergence.
The article contains no abstract
The aim of the paper is to summarize contributions of Ryszard Zieliński to two important areas of research. First, we discuss his work related to Monte Carlo methods. Ryszard Zieliński was particularly interested in Monte Carlo optimization. About 10 of his papers concerned stochastic algorithms for seeking extrema. He examined methods related to stochastic approximation, random search and global optimization. We stress that Zielinski often considered computational problems from a statistical perspective....
We consider the generalized Z^-norm optimization problem assuming that the joint probability distri-bution of random variables is unknown. The solution to the problem has, therefore, to be estimated from a sample. We examine a natural estimator and show its strong consistency and asymptotic normality under quite general assumptions. Certain discrimination and screening problems, formalized in decision- theoretical manner, can be solved using Z^-norm minimization procedures. We derive asymptotic...
The article contains no abstract
We consider some fundamental concepts of mathematical statistics in the Bayesian setting. Sufficiency, prediction sufficiency and freedom can be treated as special cases of conditional independence. We give purely probabilistic proofs of the Basu theorem and related facts.
Two-stage sampling schemes arise in survey sampling, especially in situations when the complete update of the frame is difficult. In this paper we solve the problem of fixed precision optimal allocation in two special two-stage sampling schemes. The solution is based on reducing the original question to an eigenvalue problem and then using the Perron-Frobenius theorem.
It is easy to notice that no sequence of estimators of the probability of success θ in a Bernoulli scheme can converge (when standardized) to N(0,1) uniformly in θ ∈ ]0,1[. We show that the uniform asymptotic normality can be achieved if we allow the sample size, that is, the number of Bernoulli trials, to be chosen sequentially.
Let Z1 and Z2 be observable random variables. Assume they depend on latent trait U and are conditionally independent, given U. 1) How, and to what extent, the joint distribution of (U, Z1, Z2) can be recovered from that of (Z1,Z2)? 2) Suppose that, knowing Z1 and/or Z2, we are to make decision concerning U. What decision rule is the best? Both the problems are properly formalized and solved in the simple case of binary U.
Page 1