Reliability Of Such Models Research Articles

Vapour cloud explosions — an analysis based on accidents: Part I

The accidental release of a combustible gas or liquid may result in an explosive vapour cloud which upon ignition will form a threat to the surrounding area. Models have been developed in order to quantify this effect, but still many questions regarding the accuracy and reliability of such models have to be answered. As research shows the topic to be very complicated, an alternative approach is presented in this paper. This approach is based on the accidents that happened in the past and it is presented in two parts. Part I covers the derivation of trends under which the accidents took place, whereas part II describes a comparison of accidents with a theoretical model.

An Examination of the Conceptual Issues Involved in Developing Credit-Scoring Models

Multiple discriminant analysis (MDA) is frequently used to develop statistical credit-scoring models for loan evaluation purposes. Current legislative efforts to insure that credit is being granted in a nondiscriminatory manner have focused considerable attention on the reliability of such models. This article examines the theoretical requirements of the MDA model in the context of a realistic lending situation and illustrates the extent of bias when these theoretical assumptions are not fully met. The article concludes that failure to rigorously meet all the theoretical assumptions of the statistical model may not be as critical as insuring that credit managers fully understand the limitations of these types of decision tools. Furthermore, the evidence indicates that statistical models other than multiple discriminant analysis are possibly more relevant to the credit-granting decision.

Densities from the CACTUS accelerometer as an external test of the validity of the thermospheric models

Total density data were obtained from the accelerometer CACTUS on board of CASTOR-D5B 1975-39A. Numerous and precise data were obtained between 250km and 600km altitude in the equatorial region (±30° latitude) for a period extending from May 1975 (minimum of solar activity) to February 1979 (already important solar activity). Since CACTUS data have not yet been used for the construction of empirical thermospheric models, a significant part of the data file is compared with several thermospheric models in order to provide an external test of the reliability of such models. Standard deviations of the order of 20% are apparent. The most significant differences extend over a few weeks and cannot be represented by the geophysical indices as they are presently used in the empirical models. Such an experimental fact suggests that the mathematical and physical aspects of the empirical models should be refined in order to achieve a better representation of physical reality.

Confidence Intervals and the Evaluation of Regression Based Appraisal Models

Perhaps the most important technical development in assessment administration in recent years has been the application of multiple regression analysis. The technique offers an effective means of reappraising the large majority of single family residences each year or so at low costs per parcel. This paper addresses the issue of how to evaluate properly the accuracy or reliability of such models.

Calculation of inelastic scattering in terms of elastic scattering

It is shown that the adiabatic distorted-waves theory of inelastic scattering, under suitable approximations, can be re-expressed in terms of derivatives of elastic scattering amplitudes. The methods are very general, and have the interesting aspect that they display relations which are directly among scatering amplitudes, so that the role of the optical potential is minimized. Results are obtained for all orders of the inelastic coupling constant, hence for single excitation (normal inelastic scattering), for double excitation, and for the reaction of inelastic coupling on the elastic scattering. The theory easily takes the form of a generalized “diffraction model,” and yields information about the reliability of such models. The older, Fraunhofer model of Blair is obtained as a special case. Two basic approximations are employed. Both are at their best if the projectile is absorbed strongly by the target nucleus, and if angular momentum transfers are much smaller than the angular momenta that are important for partial wave expansions. Approximation A simplifies the Green's functions that enter in calculations of the higher-order processes, so that the mathematical structure of such calculations becomes equivalent to that of first-order calculations by others which are known from the study of elastic scattering, and which are easily related to derivatives of the amplitudes for elastic scattering.