Admissible linear estimation in singular linear models

The admissibility results of Rao (1976), proved in the context of a nonsingular covariance matrix, are exteneded to the situation where the covariance matrix is singular. Admi.s s Lb Le linear estimators in the Gauss-Markoff model are characterized and admis-sibility of the best linear unbiased estimator is investigated.     

Estimation in singular partitioned,reduced or transformed linear models

A singular partitioned linear model, i.e. the singular model comprising the main parameters and the nuisance parameters, can be reduced, or transformed to the form in which only linear functions concerning main parameters are involved. In the paper some properties of the best linear unbiased estimators of these functions following from these models are considered.     

Categorical information and the singular linear model

C. R. Rao (1978) discusses estimation for the common linear model in the case that the variance matrix σ² Q has known singular form Q . In the more general context of inference, this model exhibits certain special features and illustrates how information concerning unknowns can separate into a categorical component and a statistical component. The categorical component establishes that certain parameters are known in value and thus are not part of the statistical inference.     

Estimation of parameters under a random censorship model

The problem of estimation of parameters of a lifetime distribution is considered under the proportional hazards model of random censorship. Asymptotic variances of several estimators of survival function are compared in the eponential case.     

Numerical treatment of restricted gauss-markov model 1

The singular value decomposition (SVD) has been widely used in the ordinary linear model and other statistical problems. In this paper, we shall introduce the generalized singular value decomposition (GSVD) of any two matrices X and H having the same number of columns to moti-vate the numerical treatment of large scale restricted Gauss-Markov model (y,XβHβ = r,σ²1), a situation to reveal the relationship (or restriction) existing among the parameters of the model. Many approaches to restricted linear model are already available. Those approaches apply the generalized inverse of matrices and emphasize the the-oretical solution of the problem rather than the development of efficient and numerical stable algorithm for the computation of estimators. The possible merit of the method present here might lie in the facts that they directly lead to an efficient, numerically stable and easily programmed algorithm for     

On quadratic estimation of heteroscedastic variances

Beginning with a brief introduction to the general theory the concept of Bayes invariant quadratic unbiased estimators (BAIQUEs) founded by Kleffe and Pingus(1974)is applied to combined samples with a common mean and different variances.Explicite formulas for Baique under these special assumptions are derived.Finally,some numerical comparisons of the variance function of Baiques under different prior distributions are given.     

Estimation of parameters of a pareto distribution by generalized order statistics

In this paper we address the problem of estimating the parameters of Pareto II distribution based on generalized order statistics. The estimators based on order statistics and record values are shown to be special cases of these estimators.     

Least squares theory for possibly singular models

In a recent paper, Scobey (1975) observed that the usual least squares theory can be applied even when the covariance matrix σ²V of Y in the linear model Y = Xβ + e is singular by choosing the Moore-Penrose inverse (V+XX′)⁺ instead of V^-1 when V is nonsingular. This result appears to be wrong. The appropriate treatment of the problem in the singular case is described.     

Estimation of reliability under ordered restriction on the parameters

This paper deals with the estimation of reliability for a strength-stress model under ordered restriction on the parameters. It is assumed that components have exponential distributions and are arranged in a parallel system and the failure of one component, results in increasing the failure rate of the remaining components. Results are derived when (i) the ordering of the means is taken into account and when (ii) the ordering of the means is ignored. Simulation studies are carried out to compare the results. It is noticed that, in almost all cases, in case (i) the estimates are closer to the true value with smaller mean squared error (MSE) and smaller’ standard deviation than in case (ii). Thus when the ordering of the means is present in the model, such information should be incorporated in the estimation of reliability.     

Estimation of regression vectors in linear mixed models with Dirichlet process random effects

The Dirichlet process has been used extensively in Bayesian non parametric modeling, and has proven to be very useful. In particular, mixed models with Dirichlet process random effects have been used in modeling many types of data and can often outperform their normal random effect counterparts. Here we examine the linear mixed model with Dirichlet process random effects from a classical view, and derive the best linear unbiased estimator (BLUE) of the fixed effects. We are also able to calculate the resulting covariance matrix and find that the covariance is directly related to the precision parameter of the Dirichlet process, giving a new interpretation of this parameter. We also characterize the relationship between the BLUE and the ordinary least-squares (OLS) estimator and show how confidence intervals can be approximated.     

Estimation of parameters of the inverse gaussian-weibull mixture model

The main objective of this work is to estimate the 5-dime-nsional vector of parameters (p,μ,λ,α,c) of the mixture of an Inverse Gaussian IG(μ,λ) and Weibull W(α,c) distributions with mixing proportion p. We use the maximum Likelihood method (MLM) and the weighted maximum likelihood method (WMLM), both under the sampling schemes suggested by Hosmer (1973). Simulation study shows that the WMLM performs best, when Hosmer's model 2 is used, in the sense of minimizing the mean square error.     

Linear model estimation errors with estimated design parameters

The problem of error estimation of parameters b in a linear model,Y = Xb+ e, is considered when the elements of the design matrix X are functions of an unknown ‘design’ parameter vector c. An estimated value c is substituted in X to obtain a derived design matrix [Xtilde]. Even though the usual linear model conditions are not satisfied with [Xtilde], there are situations in physical applications where the least squares solution to the parameters is used without concern for the magnitude of the resulting error. Such a solution can suffer from serious errors.

This paper examines bias and covariance errors of such estimators. Using a first-order Taylor series expansion, we derive approximations to the bias and covariance matrix of the estimated parameters. The bias approximation is a sum of two terms:One is due to the dependence between ? and Y; the other is due to the estimation errors of ? and is proportional to b, the parameter being estimated. The covariance matrix approximation, on the other hand, is composed of three omponents:One component is due to the dependence between ? and Y; the second is the covariance matrix ∑_b corresponding to the minimum variance unbiased b, as if the design parameters were known without error; and the third is an additional component due to the errors in the design parameters. It is shown that the third error component is directly proportional to bb'. Thus, estimation of large parameters with wrong design matrix [Xtilde] will have larger errors of estimation. The results are illustrated with a simple linear example.     

Some matrix results related to a partitioned singular linear model

Consider the linear model (y, Xβ V), where the model matrix X may not have a full column rank and V might be singular. In this paper we introduce a formula for the difference between the BLUES of Xβ under the full model and the model where one observation has been deleted. We also consider the partitioned linear regression model where the model matrix is (X₁: X₂) the corresponding vector of unknown parameters being (β′₁ : β′₂)′. We show that the BLUE of X₁ β₁ under a specific reduced model equals the corresponding BLUE under the original full model and consider some interesting consequences of this result.     

Estimation in semiparametric models using an auxiliary model

Two classes of semiparametric and nonparametric mixture models are defined to represent general kinds of prior information. For these models the nonparametric maximum likelihood estimator (NPMLE) of an unknown probability distribution is derived and is shown to be consistent and relative efficient. Linear functionals are used for the estimation of parameters. Their consistency is proved, the gain of efficiency is derived and asymptotical distributions are given.     

Estimation of reliability for a multicomponent survival stress-strength model based on exponential distributions

Uniformly minimum variance unbiased estimator (UMVUE) of reliability in stress-strength model (known stress) is obtained for a multicomponent survival model based on exponential distributions for parallel system. The variance of this estimator is compared with Cramer-Rao lower bound (CRB) for the variance of unbiased estimator of reliability, and the mean square error (MSE) of maximum likelihood estimator of reliability in case of two component system.     

Estimation of parameters of yamada et al. model and modification of the model

Yamada, Ohba and Osaki (1983) suggested an important NHPP model for software failure phenomenon. So far little work has been done on the problem of estimating its parameters. We present here some conditions for the likelihood estimates to be finite, positive and unique. We also suggest a modification of the model. The performance measures and statistical inferences of the modified model are discussed here. The modified model is applied to software failure data and the results are compared with Jelinski-Moranda [4] and some existing important NHPP models     

Estimation of parameters and reliability from generalized life models

Simultaneous estimation of parameters with p (≥ 2) components, where each component has a generalized life distribution, is considered under a sum of squared error loss function. Improved estimators are obtained which dominate the maximum likelihood and the niinimum mean square estimators. Robustness of the improved estimators is shown even when the component distributions are dependent. The result is extended to the estimation of the system reliability when the components are connected in series. Several numerical studies are performed to demonstrate the risk improvement and the Pitman closeness of the new estimators.     

On equalities between BLUES,WLSEs, and SLSEs

Necessary and sufficient conditions for equalities between the best linear unbiased estimator, the weighted least-squares estimator, and the simple least-squares estimator of the expectation vector in a general Gauss-Markoff model are given in some alternative formulations. The main result states, somewhat surprisingly, that the weighted least-squares estimator cannot be identical with the simple least-squares estimator unless they both coincide with the best linear unbiased estimator.     

Optimality of blue's in a general linear model with incorrect design matrix

We consider the Gauss-Markoff model (Y,X₀β,σ²V) and provide solutions to the following problem: What is the class of all models (Y,Xβ,σ²V) such that a specific linear representation/some linear representation/every linear representation of the BLUE of every estimable parametric functional p'β under (Y,X₀β,σ²V) is (a) an unbiased estimator, (b) a BLUE, (c) a linear minimum bias estimator and (d) best linear minimum bias estimator of p'β under (Y,Xβ,σ²V)? We also analyse the above problems, when attention is restricted to a subclass of estimable parametric functionals.     

Estimation of the binomial parameter,using an apriori hypothesis about its value

We study estimation of the parameter p of a r.v. x ～ Bin(p,n) using the prior hypothesis that p equals a prepecified value P_o when we entertain the possibility that P_o is not the right value of p, which then could be any value in (0.1). We apply notions of Hodges and Lehmann (1952), Bickel (1984) and Berger (1982) to obtain (biased) estimators which do well under the hypothesis that p = p_o at a small price in maximum risk. A number of examples and comparisons are discussed at the end of the paper.