A case study of expert judgment: Economists' probabilities versus base-rate model forecasts

In this case study of economists' forecasts concerning economic downturn, we examine key issues concerning the psychology of prediction and the controversy surrounding the value of expertise in forecasting. We examine when experts' knowledge promotes forecast accuracy and whether biases found in psychological studies (including underutilization of relevant base rates and tendencies to extreme prediction) occur in these economic forecasts. Experts' forecasts were compared to forecasts derived from base-rate models that relied on the historical frequencies of economic downturns. The performance patterns of the experts and models crossed over the forecast horizon. Experts outperformed models in shorter-term forecasting, whereas models outperformed experts in longer-term forecasting. These results highlight the abilities and limits of experts and models in prediction and the sources of their inaccuracy.     

Historical reliability: a key to differentiating populations among patients presenting signs of multiple personality disorder.

The clinical value of inconsistencies in the historical data of patients presenting with signs of multiple personality disorder was assessed. Three major inconsistencies in historical data were identified in 23 patients who were admitted to a Dissociative Disorders Program with a diagnosis of Multiple Personality Disorder. This group showed psychological disturbance that was quite different from 23 patients with a similar discharge diagnosis but who did not exhibit inconsistencies in history. The former group reported less psychological disturbance and fewer dissociative experiences. Patients who are less credible by virtue of inconsistencies in their historical reports may belong to a different diagnostic group. The findings point to the need to chart history carefully in authenticating genuine cases.     

A factor analysis of a 100-item fear survey inventory

A 100-item fear survey schedule was constructed by compiling all previously published fear surveys to effect a more comprehensive assessment technique. It was administered to 435 college students. Reliability estimates of this instrument approximated 0.87. An orthogonal factor analysis was performed, and yielded 21 interpretable factors within each sex. Sixteen of these were similar across sex. The major areas of fear response were: social criticism; medical intervention; contamination; sudden noises; cars; death.

Previous research had indicated a four factor structure. Results of this study indicate that the number of factors varies with the number of input items. The present study presents a more comprehensive picture of the underlying factor structure of fear responding.     

The initial capitalization superiority effect in German: evidence for a perceptual frequency variant of the orthographic cue hypothesis of visual word recognition

A perceptual frequency variant of the orthographic cue (OC) hypothesis (Peressotti, Cubelli, & Job, 2003) was tested in two perceptual identification experiments using the variable viewing position technique: German nouns and non-nouns that are most frequently perceived with or without initial letter capitalization, respectively, were tachistoscopically presented in upper-case, lower-case, or with initial capitalization. The results indicated that words were best recognized in the form they are most frequently perceived in, which suggests that during reading acquisition abstract as well as case- and item-specific OCs may be learned and used for recognition.     

Estimating long-range dependence in time series: An evaluation of estimators implemented in R

Recent studies have shown that many physiological and behavioral processes can be characterized by long-range correlations. The Hurst exponent H of fractal analysis and the fractional-differencing parameter d of the ARFIMA methodology are useful for capturing serial correlations. In this study, we report on different estimators of H and d implemented in R, a popular and freely available software package. By means of Monte Carlo simulations, we analyzed the performance of (1) the Geweke—Porter-Hudak estimator, (2) the approximate maximum likelihood algorithm, (3) the smoothed periodogram approach, (4) the Whittle estimator, (5) rescaled range analysis, (6) a modified periodogram, (7) Higuchi’s method, and (8) detrended fluctuation analysis. The findings—confined to ARFIMA (0, d, 0) models and fractional Gaussian noise—identify the best estimators for persistent and antipersistent series. Two examples combining these results with the step-by-step procedure proposed by Delignières et al. (2006) demonstrate how this evaluation can be used as a guideline in a typical research situation.