An Empirical Test of the Strategic-Grid Model of Information Systems Planning*

Contingency models of information systems planning predict that no single planning approach will suit all organizations' needs. Little empirical research has been undertaken, however, to evaluate this prediction. Accordingly, we used McFarlan, McKenney, and Pyburn's (1983) strategic-grid model to study the information systems planning problems encountered by 49 governmental agencies. Twenty-seven agencies were required to follow a planning approach best suited to organizations that had a high level of dependence on both their existing and proposed systems. We predicted that agencies not having these characteristics would encounter the most problems with the approach. The remaining 22 agencies could choose their own planning approach. We studied this latter group to determine whether the problems encountered by the first group could be attributed to the mandated approach. Overall, the empirical results obtained were equivocal. Some results indicated that more planning problems were encountered by agencies in which the mandated approach was not appropriate to their position in the strategic grid. Other results were not supportive of this proposition. More work needs to be undertaken, therefore, to evaluate the predictive and explanatory power of contingency models of information systems planning. In addition, our research indicates a need to develop more rigorous theories of information systems planning behaviors, to improve the instruments needed to measure these behaviors, to explore the relationship between information systems planning behaviours and organizational effectiveness, to investigate how organizational learning impacts planning behaviors, and to determine the types of information systems planning problems that diffuse through organizations and those that remain localized.     

Inference in Arch and Garch Models with Heavy–Tailed Errors

ARCH and GARCH models directly address the dependency of conditional second moments, and have proved particularly valuable in modelling processes where a relatively large degree of fluctuation is present. These include financial time series, which can be particularly heavy tailed. However, little is known about properties of ARCH or GARCH models in the heavy–tailed setting, and no methods are available for approximating the distributions of parameter estimators there. In this paper we show that, for heavy–tailed errors, the asymptotic distributions of quasi–maximum likelihood parameter estimators in ARCH and GARCH models are nonnormal, and are particularly difficult to estimate directly using standard parametric methods. Standard bootstrap methods also fail to produce consistent estimators. To overcome these problems we develop percentile–t, subsample bootstrap approximations to estimator distributions. Studentizing is employed to approximate scale, and the subsample bootstrap is used to estimate shape. The good performance of this approach is demonstrated both theoretically and numerically.     

An Explanation of the Occupational Structure of Sectors of Industry

Abstract. This study has been carried out by the Research Centre for Education and Labour Market (ROA) as part of a long-range research project commissioned by the Ministry of Education and Sciences. The objective of the project is to develop an education/labour-market information system to help secondary and university students choose the right type of school and/or occupation (ROA 1988). The study reported on here sets out to disclose the factors underlying the development of the occupational pattern. The shares of occupational classes in each economic sector are taken as points of departure. The explanatory variables are technological progress and cyclical and structural sectoral developments. For each occupational class, an estimation is made with the help of a linear model, the estimation method being that of Weighted Least Squares. To permit several variables to be included in the equation, the data have been pooled over sectors and time. In two fifths of all cases, a model with explanatory variables has been chosen; in the remaining instances a sector-dummy model was found to suffice.