Recurrent PSA after prostatectomy for prostate cancer: implications of PSA doubling time

After radical prostatectomy for prostate cancer, men frequently develop detectable levels of prostate specific antigen (PSA). A slow rate of increase, as characterized by the PSA doubling time (PSADT) is the principal marker for a favorable prognosis. Data and results presented in 2 recent clinical articles studying cohorts of men with clinical stage T1/T2 prostate cancer are reviewed and used to develop mortality analyses. Life-table analysis shows a mortality ratio of 257% at 5 years for Gleason score < 8, PSA recurrence > 2 years after surgery for clinical stage T1/T2 disease, and PSA doubling time (PSADT) > 10 months. Markov modeling using transition probabilities derived from the clinical articles to develop a life table analysis yields a mortality ratio of 145% at 10 years for similar patients.     

Improving decision-making with scenario planning

Decision makers struggle with four important contributors to decision failure, namely, 1) bounded rationality, 2) a tendency to consider only external variables, 3) the stickiness and friction of information and knowledge, and 4) mental models that include decision premises or policies. Whether independent or combined, these decision problems can form the ultimate in decision error—folly. The four problems in decision-making are reviewed in detail and scenario planning is posited as a tool for preventing the impact of each, ultimately with an aim of avoiding folly.     

Linear versus process tracing approaches to judgment modelling: A new perspective on cue importance

For a bankruptcy prediction problem, the judgment formation process is studied using linear models and process tracing models. The linear models are constructed using traditional linear discriminant analysis techniques. The process tracing models are constructed using computer-generated algorithmically-based decision nets. All the models presented show good predictive accuracy. However, the linear models and process tracing models diverge widely on several measures of cue importance. This divergence, for a fairly straightforward problem, is intriguing since virtually all the evidence in the accounting literature about cue importance is based on linear models research. The importance of different information cues to decision-makers is clearly a critical issue in the design of effective accounting information systems. Thus, this study suggests the need for much more careful attention to the complex question of assessing cue importance.     

The Effects of Catholic Schooling on Civic Participation

The promotion of adult civic engagement is one of the primary goals of public schools. And the putatively negative effects of private schooling on civic engagement provide one of the most fundamental motivations for publicly provided schooling. In this study, I examine the comparative effects of Catholic and public high schools on adult voter participation and volunteering in the United States. I find that students who attended Catholic high schools are actually more likely to vote, though not volunteer, as adults. These estimated effects are robust to conditioning on a rich set of individual, family and community traits. I also present two-stage least squares estimates, which provide suggestive evidence that these results are not due to selection biases.JEL Code: I21, D72, H52     

Optimal foreign trade for the price-insensitive soviet-type economy

For a centrally planned economy in which the planners do not substitute among goods, either in production or consumption, in response to changes in relative prices, the short-run offer curve is in general backward-bending, and the planners will be indifferent as to whether the country is “small” or “large” in world markets as long as it faces an elastic foreign offer curve. The Soviet Union is found to have a backward-sloping offer curve in trade with the Western market economies. This suggests that systematic Soviet manipulation of this trade in order to earn monopoly profits is unlikely.     

Random coefficient regression and the long-run phillips curve for the U.S.

This research suggests that a random coefficient regression model is well-suited for analyzing long-run versus short-run movements in wages and unemployment and is an improvement over some empirical techniques typically used.