From equivalent linear equations to Gauss-Markov theorem.
A note on correlation coefficient between random events
Correlation coefficient is a well known measure of (linear) dependence between random variables. In his textbook published in 1980 L.T. Kubik introduced an analogue of such measure for random events A and B and studied its basic properties. We reveal that this measure reduces to the usual correlation coefficient between the indicator functions of A and B. In consequence the resuts by Kubik are obtained and strenghted directly. This is essential because the textbook is recommended by many universities...
Inverting covariance matrices
Some useful tools in modelling linear experiments with general multi-way classification of the random effects and some convenient forms of the covariance matrix and its inverse are presented. Moreover, the Sherman-Morrison-Woodbury formula is applied for inverting the covariance matrix in such experiments.
On useful schema in survival analysis after heart attack
Recent model of lifetime after a heart attack involves some integer coefficients. Our goal is to get these coefficients in simple way and transparent form. To this aim we construct a schema according to a rule which combines the ideas used in the Pascal triangle and the generalized Fibonacci and Lucas numbers
Selective lack-of-memory and its application
We say that a random variable X taking nonnegative integers has selective lack-of-memory (SLM) property with selector s if P(X ≥ n + s/X ≥ n) = P(X ≥ s) for n = 0,1,.... This property is characterized in an elementary manner by probabilities pₙ = P(X=n). An application in car insurance is presented.
Characterizing experimental designs by properties of the standard quadratic forms of observations
For any orthogonal multi-way classification, the sums of squares appearing in the analysis of variance may be expressed by the standard quadratic forms involving only squares of the marginal and total sums of observations. In this case the forms are independent and nonnegative definite. We characterize all two-way classifications preserving these properties for some and for all of the standard quadratic forms.
On distribution of waiting time for the first failure followed by a limited length success run
Many doctors believe that a patient will survive a heart attack unless a succeeding attack occurs in a week. Treating heart attacks as failures in Bernoulli trials we reduce the lifetime after a heart attack to the waiting time for the first failure followed by a success run shorter than a given k. In order to test the "true" critical period of the lifetime we need its distribution. The probability mass function and cumulative distribution function of the waiting time are expressed in explicit and...
Quotient of information matrices in comparison of linear experiments for quadratic estimation
The ordering of normal linear experiments with respect to quadratic estimation, introduced by Stępniak in [Ann. Inst. Statist. Math. A 49 (1997), 569-584], is extended here to the experiments involving the nuisance parameters. Typical experiments of this kind are induced by allocations of treatments in the blocks. Our main tool, called quotient of information matrices, may be interesting itself. It is known that any orthogonal allocation of treatments in blocks is optimal with respect to linear...
Approximate sums of squares in analysis of variance
Consider the two-way crossed classification model, in which there are a levels of the factor A, b levels of the factor B and nij observations y(i,j,k), k=1,⋯,n(i,j), for the (i,j)th cell, i=1,⋯,a, j=1,⋯,b. The sum of squares for testing interactions in this model can be written as Q=∑(i,j)n(i,j)(y(i,j,⋅)/n(i,j)−y(i,⋅,⋅)/n(i,⋅)−y(⋅,j,⋅)/n(j,⋅)+y(⋅,⋅,⋅)/n(⋅,⋅))^2, where y(i,j,⋅)=∑(k)y(i,j,k), y(i,⋅,⋅)=∑(j)y(i,j,⋅), y(⋅,j,⋅)=∑(i)y(i,j,⋅), y(⋅,⋅,⋅)=∑(i)y(i,⋅,⋅), n(i,⋅)=∑(j)n(i,j), n(⋅,j)=∑(i)n(i,j)...
Towards a notion of testability
The problem of testability has been undertaken many times in the context of linear hypotheses. Almost all these considerations restricted to some algebraical conditions without reaching the nature of the problem. Therefore, a general and commonly acceptable notion of testability is still wanted. Our notion is based on a simple and natural decision theoretic requirement and is characterized in terms of the families of distributions corresponding to the null and the alternative hypothesis. Its consequences...
A sufficient condition for admissibility in linear estimation
It was recently shown that all estimators which are locally best in the relative interior of the parameter set, together with their limits constitute a complete class in linear estimation, both unbiased and biased. However, not all these limits are admissible. A sufficient condition for admissibility of a limit was given by the author (1986) for the case of unbiased estimation in a linear model with the natural parameter space. This paper extends this result to the general linear model and to biased...
Quadratic unbiased estimation of nonstandard parametric functions
Admissible invariant estimators in a linear model
Let be observation vector in the usual linear model with expectation and covariance matrix known up to a multiplicative scalar, possibly singular. A linear statistic is called invariant estimator for a parametric function if its MSE depends on only through . It is shown that is admissible invariant for , if and only if, it is a BLUE of in the case when is estimable with zero variance, and it is of the form , where and is an arbitrary BLUE, otherwise. This result is used in...
Page 1