Mark S. McNulty

Live Cattle Futures Response to Cattle on Feed Reports

Knight-Ridder surveys provide prerelease expectations of the information in the Cattle on Feed reports. The Knight-Ridder information is employed here, in accordance with the efficient markets hypothesis, to identify the unanticipated information provided in Cattle on Feed reports. Rational expectations theory is used to test Knight-Ridder forecasts for unbiasedness, efficiency, and forecast performance. The forecasts mostly satisfy rationality conditions. Live cattle futures prices respond to unanticipated information about placements and marketings contained in Cattle on Feed releases. However, report information is absorbed quickly.

An analysis of feeder cattle price differentials

A static premium and discount analysis was used to determine whether certain management or marketing practices affect the price of feeder cattle at auction. Data included animal characteristics (sex, weight, muscle, frame, horns, breed, condition, fill, health) and market characteristics (price, lot size, market location, auction sale order). The analysis shows that marketing price can be enhanced by selling heavy muscled, crossbred cattle with either medium or large frame in large (truck-load) size lots. Cattle should carry average fill, an average or slightly fleshy amount of condition, and be dehorned.

The Independent Pairs Assumption in Hypothesis Tests Based on Rank Correlation Coefficients

Abstract A simple example and a computer simulation study are given to illustrate the inflation that can occur in the Type I error when the independent pairs assumption is violated for hypothesis tests based on rank correlation coefficients. The specific types of random variables considered are those that arise from random walk, moving average, and autoregressive processes.

Stochastic models for software science

Abstract The current development of software measures is the result of treating software construction as a deterministic process. In this article, stochastic models for the software construction process are developed. The models are based on absorbing Markov chains and can be used to estimate standard software measures and their variability. The focus is on the length measure. The predictive properties of the models are investigated in an empirical study based on data from PASCAL and C language software.

Evidence on Agglomeration in Quality Space--Revisisted

This paper extends the results of C. Hjorth-Andersen (1988) by examining agglomeration in quality space for 16,257 consumer products tested by Konsument and Test magazines. These magazines have tested and evaluated quality characteristics for a large number of brands of different consumer products, thereby indicating brand positions in quality space. The results indicate that markets for consumer products cluster around average and high quality. Manufacturers tend to provide consistent attribute qualities. The results remain consistent for different product classifications and different definitions of quality space. Copyright 1993 by Blackwell Publishing Ltd.

Market equilibria with endogenous, hierarchical information

Abstract A competitive industry in which there is both stochastic demand and a production lag is examined. All agents observe costless endogenous information (prices), and informed agents also observe costly exogenous information. Solutions for market equilibria are found under quite general conditions. A number of issues are examined, including 1. 1) the way in which prices transmit information from the informed to the uninformed, 2. 2) how the specification of the private information set and of the stochastic processes driving demand influence market equilibria, 3. 3) whether the equilibrium distribution of information may expected to be hierarchical, and 4. 4) whether the equilibrium distribution of information is socially optimal.

The sample spectrum of time series with trading day variation

A time series that is constructed as the monthly total of the daily levels of some activity will contain trading day variation. This letter presents the sample spectrum of a trading day variable. The presence of the identified patterns in the sample spectrum of an observed time series indicates the presence of trading day variation.

Response to: Resolving the software science anomaly

Abstract Card and Agresti 1. [1] claim to have resolved the software science anomaly by demonstrating that Halsteads length measure lacks empirical as well as theoretical support. It may be true that software science, as developed by Halstead 2. [2],lacks this support but the arguments used by Card and Agresti are shown to be invalid.

Trading-Day Variation: Theory and Implications for Monthly Meat Demand

The authors consider the problem of fitting regression models with data containing trading-day variation. Multiplicative and additive expressions for trading-day variation are presented. Multiplicative adjustment is more reasonable than the additive but is also more complex. Expressions are derived for biases that trading-day variation introduces into least squares estimators. Estimated retail demands for beef, pork, and chicken show the biases are large and in the directions predicted when monthly data are used, but are small when quarterly data are used. Multiplicative adjustment is statistically superior to additive adjustment, although practical differences are small.

Testing for autocorreiation in nested analysis of variance

This paper considers the problem in which repeated measures are taken on each subject in an experimental design and the model errors follow a first order autoregressive process. The appropriate form of the Durbin-Watson test statistic for autocorrelated errors is presented. Critical values for the test statistic are reported.