In classical multiple linear regression analysis problems will occur if the regressors are either multicollinear or if the number of regressors is larger than the number of observations. In this note a new method is introduced which constructs orthogonal predictor variables in a way to have a maximal correlation with the dependent variable. The predictor variables are linear combinations of the...

In some physical systems, where the goal is to describe behavior over an entire eld using scattered observations, a multiple regression model can be derived from the discretization of a continuous process. These models often have more parameters than observations. We propose a technique for constructing smoothed estimators in this situation. Our method assumes the model has random explanatory a...

This paper studies estimation of multiplicative, unobserved components panel data models without imposing the strict exogeneity assumption on the explanatory variables. The method of moments estimators proposed have significant robustness properties; they require only a conditional mean assumption, and apply to models with lagged dependent variables, finite distributed lag models that allow arb...

The graphical tools (see Section 4) showed that the logarithmic transformation applied to both the response (concentration ratio) and the explanatory variable sacrifice time makes the relationship between the two variables approximately linear. In this section we will examine the relationship between the logarithm of concentration ratio (LNRATIO) and the eight explanatory variables with the fol...

Probit residuals need not sum to zero in general. However, if explanatory variables are qualitative the sum can be shown to be zero for many models. Indeed this remains true for binary dependent variable models other than Probit and Logit. Even if some explanatory variables are quantitative, residuals can sum to almost zero more often than might at first seem plausible.

We study the influence of explanatory variables in prediction by looking at the distribution of the log-odds ratio. We also consider the predictive influence of a subset of unobserved future variables on the distribution of log-odds ratio as well as in a logistic model, via the Bayesian predictive density of a future observation. This problem is considered for dichotomous, as well as continuous...

In analyses of contingency tables made up of categorical variables, the study of relationship between the variables is usually the major objective. So far, many association measures and association models have been used to measure  the association structure present in the table. Although the association measures merely determine the degree of strength of association between the study varia...

Many current approaches to statistical language modeling rely on independence a.~sumptions 1)etween the different explanatory variables. This results in models which are computationally simple, but which only model the main effects of the explanatory variables oil the response variable. This paper presents an argmnent in favor of a statistical approach that also models the interactions between ...

Many current approaches to statistical language modeling rely on independence a.~sumptions 1)etween the different explanatory variables. This results in models which are computationally simple, but which only model the main effects of the explanatory variables oil the response variable. This paper presents an argmnent in favor of a statistical approach that also models the interactions between ...