Joint tests of non-nested models and general error specifications

This paper is concerned with joint tests of non-nested models and simultaneous departures from homoskedasticity, serial independence and normality of the disturbance terms. Locally equivalent alternative models are used to construct joint tests since they provide a convenient way to incorporate more than one type of departure from the classical conditions. The joint tests represent a simple asymptotic solution to the “pre-testing” problem in the context of non-nested linear regression models. Our simulation results indicate that the proposed tests have good finite sample properties.     

The risky business of insurance pricing

The factors influencing insurance pricing decisions are assessed using the ISO product liability ratemaking files for 1980–1984. The mean loss level has a strong positive effect on manual rates and premium rates/exposure. Evidence on a variety of ambiguity measures is more mixed. As a broad generalization, risk ambiguity lowers manual rates, which may reflect exclusion of large loss outliers as being unrepresentative. Risk ambiguity tends to have a positive effect on actual pricing decisions for particular policies, especially bodily injury lines and the interactive risk-ambiguity model.     

Construction and selection of quick switching double-sampling system: sample size tightening

A new sampling inspection system involving normal and tightened plans is presented. Performance measures are developed. The average sample number of the newly developed system is compared with those of existing plans. Procedures are indicated an tables are constructed for designing the system indexed by various combinations of parameters.     

Formalization of grievance procedures: A multi-firm and industry study

This study examines how two aspects of formalization of grievance procedures affect grievance procedure effectiveness: written versus oral presentation of grievance and the level of union and management authorized to resolve grievances at the first step. Several demographic variables were controlled for: size of bargaining unit, percentage of organization represented, number of grievances filed, and industry. A sample of 46 firms, representing a broad cross-section of unionized companies in the private sector, was obtained with the help of the American Arbitration Association. Results indicate that more formal grievance procedures (e.g., written presentation) are positively related to resolution rates at the first step. The number of grievances field and the size of the bargaining unit were positively related to total resolution rates prior to arbitration. We thank the American Arbitration Association for making this study possible.     

A note on estimating regression coefficients with missing data

In this note, we consider the problem of estimating regression coefficients when there are missing observations of some explanatory variables. Following Dagenais (1973), Gourieroux and Monfort (1981), and Conniffe (1983a, 1983b), we assume auxiliary relationships exist among explanatory varibles. Several estimatprs and their interrelationships are discussed. We begin with the model of Gourieroux and

Monfort (1981)     

CORPORATE TAX INCIDENCE AND INEFFICIENCY WHEN CORPORATE AND NONCORPORATE GOODS ARE CLOSE SUBSTITUTES

An important deficiency in Harberger's [1962] model of corporate income taxation is its inability to consider both corporate and noncorporate production of the same good. Within-industry substitution has potentially major implications for both the excess burden and incidence of the corporate tax.
We analyze this within-industry substitution using a model in which each industry/sector contains corporate and noncorporate firms (with identical production functions) which produce goods that are close substitutes. The scope for considerable within-industry substitution of noncorporate for corporate capital leads to a very much larger excess burden than that in the Harberger model.     

Portmanteau test for conditional heteroscedasticity, using ranks of squared residuals

Squared residual autocorrelations have been found useful in detecting departures from linearity in time series models. This is especially the case with data exhibiting heterogeneity in variances. A rank test is proposed which is much more robust than its parametric counterpart.