Sequential point estimation of normal mean under LINEX loss function

A sequential point estimation of the mean of a normal distribution is considered under LINEX loss function. The regret of sequential procedures are obtained. Furthermore, it is shown that a sequential procedure with the sample mean as an estimate is asymptotically inadmissible. An accerelated stopping time is also considered. Received: December 1999     

Presentation Matters: Comparison of Cognitive Effects of DTC Prescription Drug Advergames,Websites, and Print Ads

Applying limited cognitive capacity theory, this study examined the cognitive effects of direct-to-consumer (DTC) advergames in comparison to those of print ads and websites. The results revealed consumers’ memories of the advertised drug brand and information about the drug and the disease it treats was the lowest in the advergame condition and highest in print. For the content elements that were centrally integrated into the advergame, however, consumer recall was the highest in the advergame condition. In addition, differential memory decay was found across media types. Memory decay was greater in the print ad condition than the other media conditions.     

Uniformly most accurate equivariant prediction limit

Summary Prediction limits are widely used for reliability problems and other related problems. The determination of prediction limits has been extensively investigated, but few optimal properties of these limits have been explored. This paper introduces a concept of uniform accuracy in order to compare equivariant prediction limits and show that the prediction limits used for the normal distribution and the exponential distribution are uniformly most accurate equivariant.     

Asymptotic second-order efficiency of three-stage procedure with a warranted confidence level

This paper considers a three-stage procedure which was proposed by Holm (1995) to yield a fixed-width confidence interval of the normal mean with a precise confidence level. The procedure is shown to be asymptotically second-order efficient. The procedure is also applied to such problems as bounded risk and hypothesis testing     

The influence of natural hazards on urban housing location

The effects of natural hazards on urban housing location are investigated, using as an example seismic damage. The seismic damage function is allowed to arbitrarily vary in two-dimensional (r, θ) space rather than the usual one-dimensional models of Mills, Muth, Alonso, etc. Damage is assumed to only affect capital inputs (i.e., the structure), not land. Under the influence of expected damage, the population and CBD (treated as a point) are seen to shift toward lower expected damage. It is found that the relative importance of damage increases as the capital productivity parameter (b) of the Cobb-Douglas production function increases.