A two-stage proxy variable approach to estimating movie box office receipts

This paper advances the ongoing discussion of methods for predicting movie box office revenues with two contributions to the methodology and an out-of-sample test of the model. The first innovation is the development of a two-stage model using publicly available pre-release indicators to predict (1) initial week and (2) subsequent run box office revenues. To incorporate the experience-good nature of movies, the second stage is estimated by incorporating a proxy variable for box office success during the first week relative to predicted first week success. The second contribution is an empirical test of De Vany and Walls’ (J Econ Dyn Control 28:1035–1057, 2004) finding that the distribution of movie revenues has “heavy tails” and follows a non-Gaussian stable distribution with infinite variance. We estimate the two-stage model of a movie’s box office success on all general release movies in 1 year with both the Gaussian and stable distribution with heavy tails and infinite variance and find no evidence for the stable distribution in either stage of the estimation. This two-stage model is validated by comparing all general release movies in 3 future years (out-of-sample data) to the model’s predictions.     

Association between self-reported health and physical and/or sexual abuse experienced before age 18

OBJECTIVE: The present study evaluated the association between women's health and physical and sexual abuse suffered before age 18. METHODS: A total of 3,568 randomly sampled insured women ages 18-64 completed a telephone interview to assess history of physical only, sexual only, or both physical and sexual abuse before age 18 (Behavioral Risk Factor Surveillance System); and current health (Short Form-36, Center for Epidemiologic Studies-Depression, Presence of Symptoms surveys). Adjusted analyses compared the health of women with physical abuse only, sexual abuse only, or physical and sexual abuse to the health of women without these abuse histories. RESULTS: Poorest health status was observed in women with a history of both physical and sexual child abuse compared to women without these abuse histories. In models that adjusted for age and income, women with both abuse types had increased prevalence of depression (prevalence ratio, 2.16), severe depression (PR, 2.84), physical symptoms (PR range, 1.33 for joint pain to 2.78 for nausea/vomiting), fair/poor health (PR, 1.84), and lower SF-36 scores (3.15-5.40 points lower). Women with physical abuse only or sexual abuse only also had higher prevalence of symptoms and lower SF-36 scores but the associations were not as strong. CONCLUSIONS: This study adds to the literature showing a graded association between multiple adverse events in childhood and adult health.     

The influence of partner cues on the extinction of causal judgments in people

Studies in laboratory animals have shown that the extinction of a conditioned stimulus, A, is regulated by the associative history of a second stimulus, X, when the two are extinguished in simultaneous compound: An inhibitory X protects A from extinction (Rescorla Learning & Behavior, 31, 124–132, 2003), whereas an excitatory X facilitates, and under some circumstances deepens, the extinction of A (Rescorla Journal of Experimental Psychology: Animal Behavior Processes, 26, 251–260, 2000, Journal of Experimental Psychology: Animal Behavior Processes, 32, 135–144, 2006). In the present study, we used the allergist task to examine whether the extinction of causal judgments in people is similarly regulated by the causal status of co-present stimuli. Experiment 1 showed that a cue trained as a conditioned inhibitor protected a target cue from extinction: The target extinguished in compound with the inhibitor was rated as being more causal of the outcome than was a target extinguished in compound with a control cue lacking inhibitory properties. In contrast, the remaining experiments showed that the extinction of a target cue was regulated by the presence, but not the causal status, of a partner cue: Target cues extinguished in compound were protected from extinction, and no evidence showed that an already extinguished partner conferred more protection (Exp. 2), or that an excitatory partner conferred any less protection (Exps. 2 and 3), or that an excitatory partner deepened the extinction of its already extinguished target. These findings are inconsistent with elemental models that rely on a common error term to explain associative changes in extinction. They are largely, but not completely, consistent with the configural model proposed by Pearce (Psychological Review, 94, 61–73, 1987), which predicts an ordering of levels of protection that was not observed.     

The concept of resilience with respect to indicating instruments

This paper deals particularly with those instruments of the index-and-scale and value-controlling types of the class of non-integrating instruments, as distinguished from integrating instruments and those used for comparison purposes strictly. Independent corroboration is adduced regarding the general characteristics of the hysteresis phenomena discussed in the author's earlier paper on the variance of measuring instruments.It is shown that the area of the hysteresis loop obtained on complete cyclic calibration is a measure of the energy dissipated in the operation of the instrument and that the smallness of the area of this loop, which may be used as a factor in a function exactly analogous to the resiliency in the case of other quasielastic bodies, is a measure of the excellence of the instrument as regards the reproducibility or invariance of its indications, so far as concerns mechanical sources of variation. The specific properties of the hysteresis loop are set down analytically and the physical nature and causes of the imperfect resilience of instruments are discussed in detail.Since the motion of an index or pointer through a displacement implies the existence of a motive force to bring about the deflection, in the presence of an equivalent reactive or restoring force opposing it, the essentials of a work diagram apparently always exist in the results of a properly perforined cyclic calibration. Indications are given of the methods to be followed in the process of reducing the results of the calibration to such form that the loop obtained correctly represents energy dissipation during a cycle, typifying an integral of the form

$\text{∫}\text{B}{}_{\mathrm{y}}\text{D}{}_{\mathrm{y}}\text{[Ø}{}_1\text{(Y)?Ø}{}_2\text{(Y)] dy}$

when B_y and D_y are the projections of the extremities B and D of the loop on the, axis of deflection or reading.The restrictions which surround the carrying out of cyclic calibrations in a manner calculated to obtain results of the character required are discussed, including the requirements of slow, aperiodic change of the variables, unreversed between the chosen extreme turning points, and accomplished in the absence of jarring or vibration. Attention is directed to the necessity of accustoming the instrument to the particular cycle over which it is to be calibrated, to the end of regularizing its performance.Both the form and area of the hysteresis loop should be observed, in order to arrive at regional as well as aggregate effects of the causes producing the variance. The amount and scope of the effects of vibratory treatment in modifying and diminishing the causes of lag are indicated.The possible causes of the lag known to exist in. the case of instruments using a surface of discontinuity between fluids as the indicating element are discussed, and it is shown that known phenomena perhaps hitherto unrecognized in their relation to instrument design and calibration may account for the variancy noted. Detailed experimental consideration of these factors is hoped for.Comparison of instrument performances on the basis of the hysteresis loss requires equivalence of the ranges of operation, or possibly reduction of the results on the basis of information not vet available in a form, capable of general application.The resiliency determination opens up a field for type-testing and selection of instruments on a basis quite discrete from that pertaining to the ordinary methods of calibration, in that the former permits selection between types of instruments rather than between individuals, making clearly discriminable the differences between the characteristics of given operating principles, designs of mechanical details, or qualities of workmanship. The more general methods of diminishing variance educible from the foregoing considerations are indicated.     

A process for the critical analysis of instructional theory

Some have argued for a common language in the field of instructional design in an effort to reduce misunderstandings and simplify a multitude of synonymous terms and concepts. Others feel that this goal is undesirable in that it precludes development and flexibility. In this article we propose an ontology-building process as a way for readers to compare and analyze terms and concepts across theories. This process entails the development of categories that emerge from the literature, and the comparison of theories across categories. Such a process can reveal broader concepts that exist beyond specific theoretical terminology, differences in meanings behind common terms used by theorists, a greater understanding of the theorists’ intent, and discontinuities and gaps within the theoretical literature.