Zur Entscheidung y>y0 aufgrund der Messung einer korrelierten Variablen

Zusammenfassung Es wird eine optimale Strategie im Sinne des minimalen erwarteten Verlustes für die beiden Entscheidungeny>y _o undyy _o aufgrund der Messungen einer mitY positiv korrelierten, einfacher und/oder billiger zugänglichen ZufallsvariablenX abgeleitet. Dabei wird angenommen, daßX undY nach einer bivariaten Normalverteilung mit bekannten Parametern verteilt sind und die Entscheidungyy _o getroffen wird, wennx größer ist als ein zu bestimmendesx _o, und die Entscheidungy>y _o, wennx gleich oder kleiner als diesesx _o ist. Für die Bestimmung des optimalenx _o werden zunächst die Kosten für die beiden Fehlentscheidungen jeweils als konstant vorausgesetzt, in einem weiteren Ansatz wird jedoch für die Mißklassifikationyy _o eine mity exponentiell wachsende Risikofunktion angenommen. Um die relative Häufigkeit der zu erwartenden Fehlklassifikationen abschätzen zu können, wird schließlich die bedingte WahrscheinlichkeitP(x>x _o,y) errechnet.

---

Summary An optimal strategy, with minimum expected risk, for the decisionsy>y _o oryy _o is constructed on the basis of the measurement of a variableX, which is positively correlated withY and can be measured more easily and/or with smaller expense. A bivariate normal distribution with known parameters is assumed forX andY. For the observationsx a limitx _o is aimed at, so that the decisionsy>y _o oryy _o are taken ifx>x orxx _o respectively. Optimal values ofx _o are first calculated under the assumption of constant losses for the two misclassifications (x>x _o ifyy _o andxx _o ify>y _o). In a further approach the loss for a wrong decisionyy _o is assumed to increase exponentially withy. Finally the conditional probabilityP (x>x _o\y) is calculated to get an assessment of the relative frequencies of wrong decisions to be expected.

     

The efficiency of decomposition vs direct solution with implications for the question of decentralized or centralized planning

Conclusion The investigation provides overwhelming evidence that decomposition tends to become increasingly more efficient than direct solution as the number of subproblems becomes greater.Further investigation would be required to obtain a comparable picture of the behaviour of decomposition with respect to an increasing number of variables in the master problem as their range of variation in the present study is rather restricted.It is gratefully acknowledged: that in implementing the decompositional procedure on the computer I have benefitted from Mr. A. C. McKay's advice on matters relating to the computer programmes; also that I enjoyed the pleasant and friendly co-operation of the staff of the computer Centre, University of Birmingham, without which the difficulties involved in carrying out the present task would have been compounded manifold; that Dr. C.-L. Sandblom carefully studied this paper and suggested a number of valuable improvements. Finally, I should like to thank Professor Kronsjö for the encouragement that he has given to me in executing this project and to the research council which made this investigation possible.     

A curtailed test for the shape parameter of the Weibull distribution

Summary A procedure is proposed in this paper for testing the shape parameter, of the Weibull distribution. The test statistic which is based on the extremal quotient, possesses a monotone property which makes it possible for rejection earlier than the last planned observation of the null hypothesis,H ₀: =₀ when the alternative hypothesis isH _a: <₀ and early acceptance ofH ₀ whenH _a: >₀. The test being scale-free, does not require the scale parameter to be known.     

Developing a tutoring module for presentation of computerized patient management problems

The simulation problem is one of the most commonly used formats in computer-assisted instruction (CAI). In medical education, a patient simulation problem is known as a patient management problem (PMP). A number of computerized PMP systems are currently available. However, to date, there is no record of PMP systems having been developed using artificial intelligence. This is noteworthy since artificial intelligence techniques could help in the generation of more effective and intelligent instructional systems.

This paper analyzes and evaluates existing instructional simulation software in clinical medicine in terms of effective educational design attributes. Components of effective PMP systems thus identified are incorporated into a computerized instructional system which uses artificial intelligence techniques for teaching problem-solving and diagnostic skills to undergraduate medical students.     

FLEXIBLE MANPOWER RESOURCING: A LOCAL LABOUR MARKET SURVEY

This article presents the results of an establishment level investigation carried out late in 1986 and early in 1987 of the extent of flexible manpower resourcing in the Aberdeen local labour market. It is necessary to place these results in the context of the general background of an increasing interest in flexible manning and of certain theoretical developments in labour and industrial economics. The results suggest that whilst a variety of means are used to attain flexibility in manning, few establishments can be regarded as truly ‘flexible firms’ and manpower strategies tend to be largely reactive and ad hoc.     

An optimum sampling design to base HT-method of estimating a finite population total

The problem of constructing a sampling design with the value of the sum of second order inclusion probabilities attaining its lower bound for non-integral values of the expected effective size of a sample in the design has been considered in this paper. If the values of the characteristic of interest on all the units in the population are non-negative the design is admissible (in the sense of variance) with respect to Horvitz-Thompson estimator in the class of designs with the same set of values of the first order inclusion probabilities of the units. Again such a design is best to use Horvitz-Thompson estimator of population total in the sense of smallest average variance of the estimator under a special superpopulatio model.     

Social distance and shopping behavior

This research was designed to see whether a reconceptualization of social distance (as measured by social class variables) between shoppers and stores would provide a basis for understanding store avoidance behavior. A survey of shoppers classified into two distinct social classes investigated their images, including social class images, of two stores that had been selected as representatives of upscale and downscale retailers. A new measure of social distance was proposed and evaluated. Controlling for other functional aspects of store image such as price and personnel, the social distance measure was found to be influential in discriminating shopping frequency groups for the two stores, providing evidence that people tend to avoid stores that are perceived as being socially distant from themselves.     

The effect of interstate highway on driving times and regional shopping center drawing power

Previous studies of the driving times of shopping center patrons have not examined the effect of limited access highways. This study analyzed the impact of interstate highways on the driving times of 204 shoppers at three medium-sized regional shopping centers. Those shoppers indicating usage of interstate highways to reach their shopping destinations were significantly more likely to incur travel times greater than fifteen and twenty minutes than those not using the available interstate roads. In predicting the drawing power of these shopping centers, factors besides mass were found to be useful. These include square footage of shopping goods, stores, and number of parking places.     

On the conceptualization of theoretical constructs

An attempt is made to bridge the gap between the theorists and modellists approaches to the conceptualization of theoretical constructs. The former approach assumes that every theoretical construct has a surplus meaning over and above its operationalization and measurement as a concept. The latter approach denies this: it only accepts theoretical constructs as functions of relations between observations, i.e. measured concepts.This gap can be bridged by explicating the definition of a theoretical construct before making attempts to operationalize and measure it. An explicated definition should cover all relevant aspects (or: facets) of a construct, systematically brought together in an ideal type model. On this basis, operationalization of the construct can be directly linked to model elements, so that any possibly remaining surplus value is made explicit. The distance between theoretical construct and the measured concept can then be made as small as seems feasible; in principle, the distance can be made zero. The manner in which conservative ideology is conceptualized in a longitudinal research project in the Dutch electorate is presented as an example of this strategy.