Small sample properties of asymptotically equivalent tests for autoregressive conditional heteroskedasticity

Models that allow for autoregressive conditional heteroskedasticity (ARCH) in the error process have recently found widespread application. The purpose of this paper is to evaluate, through Monte Carlo analysis, the small sample properties of an exact Lagrange multiplier test for the presence of ARCH, and to compare the power of this test with that of an asymptotically equivalent TR² version. The comparison involves first-and higher-order variants of these processes. The results indicate substantial power differentials in favor of the exact LM test, by up to 15% for sample sizes smaller than 100.     

Forecasting expenditure on capital projects

S-curves are widely used for planning, forecasting and control of cost, time and resources of a project. In this paper, a comparison of two S-curve models developed at the Department of Health and Social Security (DHSS) and Bradford University is carried out both from the viewpoint of predictive accuracy and ease of use. The models are validated using expenditure data for 21 recent U.K. health building projects. Methods of least squares is used to estimate the parameters of the two models. These parameters are categorized according to the total cost of the projects. Both the models are shown to be of comparable accuracy for fitting actual expenditure data. The DHSS model has a major advantage of simplicity of form and use, although the slightly greater mathematical complexity of the Keller-Singh model is off-set by the readily interpretable nature of its form and basic parameters. It is concluded that both or either of the models could be used by clients/contractors for effective planning and control of project costs.     

The Person Behind The Word: Mead's Theory Of Universals And A Shift Of Focus In Symbolic Interactionism

Careful consideration of Mead's theory of universals proves to be a corrective to a number of tendencies evident in the work of at least some influential contemporary symbolic interactionists. Mead avoids any realistic hypostatization of separate universals (“meanings,” “forms,” or “essences” while at the same time eschewing nominalistic and conventionalistic views of language. His principle of the objective reality of perspectives (1932:161-175) allows him to grant objective reality to universal characteristics of concrete objects, but entails neither hypostatization nor idealization of the universal. In addition to contravening nominalism and conventionalism, Mead's theory of universals provides a perspective from which the reality of the self, and the importance of intentional action for the development of a firmly felt sense of self and autonomy, can both be affirmed. Far from being an illusion or mer.e symbolic construct, the self is seen to be an objectively real universal within the perspective of social action.     

EVALUATING FAMILY THERAPY: DIVERGENT METHODS, DIVERGENT FINDINGS

This study reports on a triangulation strategy for assessing family interaction, involving family members, their therapist and coders independently viewing videotapes. Utilizing a standardized scale, the Beavers-Timberlawn Model of Family Competence, the study found weak agreement between paired assessments within the family triad, and within the therapist-coder dyad. In addition, more complex scaling techniques such as composite "family scores" or discrepancy scores between family member dyads added no predictive power. The findings suggest that a "limit of concordance" may exist when comparing varying raters' assessments of a given family, and that methodological and/or scaling strategies designed to maximize agreement may be both fruitless and diversionary.     

Mixture models in measurement error problems, with reference to epidemiological studies

Summary. The paper focuses on a Bayesian treatment of measurement error problems and on the question of the specification of the prior distribution of the unknown covariates. It presents a flexible semiparametric model for this distribution based on a mixture of normal distributions with an unknown number of components. Implementation of this prior model as part of a full Bayesian analysis of measurement error problems is described in classical set-ups that are encountered in epidemiological studies: logistic regression between unknown covariates and outcome, with a normal or log-normal error model and a validation group. The feasibility of this combined model is tested and its performance is demonstrated in a simulation study that includes an assessment of the influence of misspecification of the prior distribution of the unknown covariates and a comparison with the semiparametric maximum likelihood method of Roeder, Carroll and Lindsay. Finally, the methodology is illustrated on a data set on coronary heart disease and cholesterol levels in blood.     

Planning Replacement of Natural Gas Distribution Systems Under Constraints on Acceptable Risk from Explosions

Natural gas distribution systems in the United States were developed primarily in the first half of this century, utilizing materials such as cast iron and then steel. Over time, cast iron and steel pipe sections became weak from corrosion and are subject to failure which in turn can lead to explosions and possible injury and loss of life. Gas utilities maintain system integrity through repair-replacement programs where pipe sections are prioritized for replacement in any given year through cost-benefit analysis; however, the total annual amount to be budgeted for replacement is left to engineering judgment. This approach has left some utilities vulnerable to criticism that their current replacement rate on cast iron pipe is not great enough and that public safety is being compromised. This paper addresses the problem situation by formulating a linear programming replacement decision model which augments cost-benefit analysis with explicit constraints on acceptable risk to human life from fire/explosion. The model is illustrated for a hypothetical utility.     

A latent Markov model for detecting patterns of criminal activity

Summary.  The paper investigates the problem of determining patterns of criminal behaviour from official criminal histories, concentrating on the variety and type of offending convictions. The analysis is carried out on the basis of a multivariate latent Markov model which allows for discrete covariates affecting the initial and the transition probabilities of the latent process. We also show some simplifications which reduce the number of parameters substantially; we include a Rasch-like parameterization of the conditional distribution of the response variables given the latent process and a constraint of partial homogeneity of the latent Markov chain. For the maximum likelihood estimation of the model we outline an EM algorithm based on recursions known in the hidden Markov literature, which make the estimation feasible also when the number of time occasions is large. Through this model, we analyse the conviction histories of a cohort of offenders who were born in England and Wales in 1953. The final model identifies five latent classes and specifies common transition probabilities for males and females between 5-year age periods, but with different initial probabilities.     

Latent variable techniques for categorical data

Two useful statistical methods for generating a latent variable are described and extended to incorporate polytomous data and additional covariates. Item response analysis is not well-known outside its area of application, mainly because the procedures to fit the models are computer intensive and not routinely available within general statistical software packages. The linear score technique is less computer intensive, straightforward to implement and has been proposed as a good approximation to item response analysis. Both methods have been implemented in the standard statistical software package GLIM 4.0, and are compared to determine their effectiveness.