H. Dennis Oberhelman

A stationary stochastic parameter model and OLS estimation of beta: A simulation study

Abstract A procedure that explicitly estimates the form of beta nonstationary followed by a security is examined. Beta estimates and forecasts generated from this process are compared with estimates and forecasts generated by the ordinary least squares (OLS) procedure. With the true underlying parameters known in the simulations, estimator basis adn efficiency are both measured. The OLS estimators and predictors are found to be quite robust, generally providing accurate estimates even when OLS conditions are seriously violated. The expanding advocacy of random-coefficient and other stochastic parameter models is open to question.

Estimating individual cross-section coefficients from the random coefficient regression model

Marketing researchers frequently encounter cross-sectional, time-series data when developing sales response models. One approach to analyzing such data is to estimate a separate OLS equation for each cross-section. Alternatively, one could pool the data from all cross-sections to estimate a single set of response coefficients for all cross-sections. However, when data are pooled, the responsiveness of individual cross-sections cannot be evaluated. In this note, we introduce a version of the random coefficient model that can be used to estimate separate sets of response coefficients for each cross-section, thereby circumventing the assumption that coefficients are homogeneous in all cross-sections. We demonstrate this approach with an empirical model that relates brand level sales to price and advertising.

Alternative tests for heteroscedasticity of disturbances:a comparative study

This paper presents three small sample tests for testing the heteroscedasticity among regression disturbances. The power of these tests are compared with two of the leading tests for this hypothesis, one by Goldfeld and Quandt [5] and the other by Theil [17]. We also provide a heuristic method of selecting the number of middle observations to be deleted for the Goldfeld-Quandt type of tests.

Yield sensitivities of short–term securities

I n recent years, short-term securities have offered extraordinarily high yields. In response, many individual investors through the assistance of money market mutual funds have moved into the market for current debt, and many corporations have devoted more resources to the management of cash. The objects of this new interest have been the prominent marketable securities: Treasury bills, bankers’ acceptances, certificates of deposit, commercial paper, and federal agency notes. The surge in popularity of these media makes it important, at this time, to assess the variability in these yields and to determine whether the pattern of variation differs among the assets and across time. The goal of this paper is to explore a major form of that variability: the systematic risk or sensitivity to changes in the rate on Treasury bills of these other four assets.3 The analysis will pertain to the monthly movement in (annualized) yields in the period from 1971 to 1981, when the yields varied greatly and reached high peaks and when the money market funds enjoyed their explosive growth. The results of this study should be of special interest to managers of cash and to those who invest in money market funds. The information presented here will permit these investors to gauge the volatility of the market value of their portfolios and to anticipate any impact on that volatility that may arise from internal shifts in the composition of the portfolios. The results of the analysis will also allow investors to understand how the volatility of their investments may change with economic conditions and/or government monetary policy.