Looking back at looking forward: A check of some fortune forecasts

We will have to wait to see how good our present forecasting methodologies are, but we can test futurologists of earlier times. The author has checked a selection of Fortune's forecasts as far back as 1933. Examination reveals that predictions were on target in three areas-technology, general trends, and structural change in the economy. They fell far short of the mark, however, in the areas of legal, political, and international constraints; the personal impact of economic factors; and the personal impact of various innovations. But the failures were often not too far off and the correct ones, if heeded, could have saved us time and money. Unfortunately, the people problems that plagued the Fortune forecasts are still unsolved today.     

Message repetition,experience and motivation

A recently proposed model describing four levels of involvement is used to predict the likely effects of repetition on subjects with high or low levels of prior experience and motivation. The predicted hypotheses are tested in an experiment, and the results are discussed in terms of their implication to communication strategy and to the recently proposed levels of involvement model.     

The Effects of Uncertainty on Market Prices: Evidence from Coaching Changes in the NBA

Abstract:  This study examines the effects of uncertainty on market prices. Specifically, we use the National Basketball Association betting market to examine whether uncertainty resulting from midseason coaching changes affects the ability of bettors to accurately set betting lines. We find that uncertainty amid midseason coaching changes results in less accurate pricing, as evidenced by higher volatility and greater overall inaccuracy in betting lines. We also find that uncertainty regarding the ability of the replacement coach and/or his strategies results in less accurate pricing, again evidenced by higher volatility and greater overall inaccuracy in betting lines.     

Efficient high-dimensional importance sampling

The paper describes a simple, generic and yet highly accurate efficient importance sampling (EIS) Monte Carlo (MC) procedure for the evaluation of high-dimensional numerical integrals. EIS is based upon a sequence of auxiliary weighted regressions which actually are linear under appropriate conditions. It can be used to evaluate likelihood functions and byproducts thereof, such as ML estimators, for models which depend upon unobservable variables. A dynamic stochastic volatility model and a logit panel data model with unobserved heterogeneity (random effects) in both dimensions are used to provide illustrations of EIS high numerical accuracy, even under small number of MC draws. MC simulations are used to characterize the finite sample numerical and statistical properties of EIS-based ML estimators.