Graphical Techniques for Selecting Explanatory Variables for Time Series Data

Bayesian model building techniques are developed for data with a strong time series structure and possibly exogenous explanatory variables that have strong explanatory and predictive power. The emphasis is on finding whether there are any explanatory variables that might be used for modelling if the data have a strong time series structure that should also be included. We use a time series model that is linear in past observations and that can capture both stochastic and deterministic trend, seasonality and serial correlation. We propose the plotting of absolute predictive error against predictive standard deviation. A series of such plots is utilized to determine which of several nested and non-nested models is optimal in terms of minimizing the dispersion of the predictive distribution and restricting predictive outliers. We apply the techniques to modelling monthly counts of fatal road crashes in Australia where economic, consumption and weather variables are available and we find that three such variables should be included in addition to the time series filter. The approach leads to graphical techniques to determine strengths of relationships between the dependent variable and covariates and to detect model inadequacy as well as determining useful numerical summaries.     

Minimum contrast estimation of random processes based on information of second and third orders

This paper presents the results on consistency and asymptotic normality of a class of minimum contrast estimators for random processes with short- or long-range dependence based on the second- and third-order cumulant spectra. Asymptotic properties of sample spectral functionals of second and third orders, which are of independent interest in view of their possible use for nonparametric estimation of processes with short- or long-range dependence, are also provided.     

Binomial and Poisson probabilities with an HP-28S scientific advanced calculator ∗

Hewlett Packard's Advanced Scientific Calculator HP-28S has built in functions to compute the CDF's of Normal, Chi-square, t and F distributions. The paper shows how to use the HP-28S calculator to obtain the CDF of the Binomial and Poisson random variables using the built in functions UTPF and UTPC of HP-28S. Illustrative examples to obtain the exact confidence intervals of the Binomial and Poisson parameters with HP-28S are given.     

Specifying and facilitating family involvement in services to persons with developmental disabilities

This paper recognizes the important role of family involvement in services to people with developmental disabilities. The paper presents a conceptualization of involvement that was an outcome of a qualitative study of involvement from the perspective of family members themselves. The authors identify a framework of family involvement the composition of which incorporates four system levels (i.e. the individual, program/agency, community, society) and the five role opportunities for family members (i.e. treatment agent, planner, advocate, evaluator, and consultant/educator). Barriers to involvement are then delineated and the authors autline a process for promoting family involvement: 1) contracting for involvement; 2) developing a support system for involvement; and 3) monitoring and evaluating involvement.Presented at the Fourth Annual Conference of The Michigan Chapter, National Association of Social Workers, May 14, 1988. Grand Rapids Michigan.The writing of this paper was supported in part by a research stimulation grant awarded to the authors by Wayne State University. Portions of this article are included in an extensive research report in Epstein, I and Grasso, A. Editors,Research Utilization in Social Work, Wayne State Univesity Press, in Press.     

Two-sided tests for change in level for correlated data

The classical problem of change point is considered when the data are assumed to be correlated. The nuisance parameters in the model are the initial level μ and the common variance σ². The four cases, based on none, one, and both of the parameters are known are considered. Likelihood ratio tests are obtained for testing hypotheses regarding the change in level, δ, in each case. Following Henderson (1986), a Bayesian test is obtained for the two sided alternative. Under the Bayesian set up, a locally most powerful unbiased test is derived for the case μ=0 and σ²=1. The exact null distribution function of the Bayesian test statistic is given an integral representation. Methods to obtain exact and approximate critical values are indicated.     

Some tests for detection of faulty inspection

Some heuristic tests for detection of existence of errors in inspection are proposed. Some of these tests are only effective if the sampling fraction ((sample size)/(lot size)) is rather large, and in all cases their application predicates special experiments to provide the requisite data. Feasibility of these experiments will vary according to specific circumstances.     

Global Trade Models and Economic Policy Analyses: Relevance,Risks and Repercussions for Africa

Computable general equilibrium models are widely used for trade policy analyses and recommendations. There is, however, increasing discomfort with the use of these models, especially in Africa. This article demonstrates that the results of several such studies of the impact of trade reforms in Africa differ drastically in terms of both magnitude and direction, failing to take account of key features of African economies. It also outlines potential consequences of the misuse of CGE models for policy evaluation and suggests pitfalls to be avoided.     

INFERENCE USING QUALITATIVE AND QUANTITATIVE INFORMATION WITH AN APPLICATION TO MONETARY POLICY

I propose a framework for drawing inferences about an unobserved variable using qualitative and quantitative information. Using this framework, I study the timing and persistence of monetary policy regimes and compute probabilistic measures of the qualitative indicator's reliability. These estimates suggest that (1) it is over one and one-half times more likely that monetary policy is not restrictive at any point in time, (2) Boschen and Mills's [1995] policy index is a reliable indicator of the stance of monetary policy, and (3) certain qualitative indicators of monetary policy improve interest rate forecasts that are based on linear forecasting models. (JEL C22, E52)