Random effects regression mixtures for analyzing infant habituation

Random effects regression mixture models are a way to classify longitudinal data (or trajectories) having possibly varying lengths. The mixture structure of the traditional random effects regression mixture model arises through the distribution of the random regression coefficients, which is assumed to be a mixture of multivariate normals. An extension of this standard model is presented that accounts for various levels of heterogeneity among the trajectories, depending on their assumed error structure. A standard likelihood ratio test is presented for testing this error structure assumption. Full details of an expectation-conditional maximization algorithm for maximum likelihood estimation are also presented. This model is used to analyze data from an infant habituation experiment, where it is desirable to assess whether infants comprise different populations in terms of their habituation time.     

Price bias and common practice in option pricing

Generally, the semiclosed-form option pricing formula for complex financial models depends on unobservable factors such as stochastic volatility and jump intensity. A popular practice is to use an estimate of these latent factors to compute the option price. However, in many situations this plug-and-play approximation does not yield the appropriate price. This article examines this bias and quantifies its impacts. We decompose the bias into terms that are related to the bias on the unobservable factors and to the precision of their point estimators. The approximated price is found to be highly biased when only the history of the stock price is used to recover the latent states. This bias is corrected when option prices are added to the sample used to recover the states' best estimate. We also show numerically that such a bias is propagated on calibrated parameters, leading to erroneous values. The Canadian Journal of Statistics 48: 8–35; 2020 © 2019 Statistical Society of Canada     

BayesProject: Fast computation of a projection direction for multivariate changepoint detection

Statistics and Computing - This article focuses on the challenging problem of efficiently detecting changes in mean within multivariate data sequences. Multivariate changepoints can be detected by...     

Ethical Numbers: Ethics Training in U.S. Graduate Statistics Programs, 2013–2014

As important members of research teams, statisticians bear an ethical responsibility to analyze, interpret, and report data honestly and objectively. One way of reinforcing ethical responsibilities is through required courses covering a variety of ethics-related topics at the graduate level. We assessed ethics requirements for graduate-level statistics training programs in the United States for the 2013–2014 academic year using the websites of 88 universities, examining 103 biostatistics programs, and 136 statistics degree programs. We categorized programs’ ethics training requirements as required or not required. Thirty-one (35.1%) universities required an ethics course for at least some degree students. Sixty-two (25.5%) degree programs required an ethics course for at least some students. The majority (77.4%) of required courses were worth 0 or 1 credit. Of the 177 programs without an ethics requirement, 19 (10.7%) listed an ethics elective. Although a single ethics course is insufficient for instilling an ethical approach to science, degree programs that model expectations through coursework point to the value of ethics in science. More training programs should prepare statisticians to consider the ethical dimensions of their work through required coursework. Supplementary materials for this article are available online.     

Media Disaster Reporting Effects on Public Risk Perception and Response to Escalating Tornado Warnings: A Natural Experiment

Previous research has evaluated public risk perception and response to a natural hazards in various settings; however, most of these studies were conducted either with a single scenario or after a natural disaster struck. To better understand the dynamic relationships among affect, risk perception, and behavioral intentions related to natural disasters, the current study implements a simulation scenario with escalating weather intensity, and includes a natural experiment allowing comparison of public response before and after a severe tornado event with extensive coverage by the national media. The current study also manipulated the display of warning information, and investigated whether the warning system display format influences public response. Results indicate that (1) affect, risk perception, and behavioral intention escalated as weather conditions deteriorated, (2) responses at previous stages predicted responses at subsequent stages of storm progression, and (3) negative affect predicted risk perception. Moreover, risk perception and behavioral intention were heightened after exposure to the media coverage of an actual tornado disaster. However, the display format manipulation did not influence behavioral responses. The current study provides insight regarding public perception of predisaster warnings and the influence of exposure to media coverage of an actual disaster event.     

Developing a Broadly Applicable Measure of Risk Perception

Decades of research identify risk perception as a largely intuitive and affective construct, in contrast to the more deliberative assessments of probability and consequences that form the foundation of risk assessment. However, a review of the literature reveals that many of the risk perception measures employed in survey research with human subjects are either generic in nature, not capturing any particular affective, probabilistic, or consequential dimension of risk; or focused solely on judgments of probability. The goal of this research was to assess a multidimensional measure of risk perception across multiple hazards to identify a measure that will be broadly useful for assessing perceived risk moving forward. Our results support the idea of risk perception being multidimensional, but largely a function of individual affective reactions to the hazard. We also find that our measure of risk perception holds across multiple types of hazards, ranging from those that are behavioral in nature (e.g., health and safety behaviors), to those that are technological (e.g., pollution), or natural (e.g., extreme weather). We suggest that a general, unidimensional measure of risk may accurately capture one's perception of the severity of the consequences, and the discrete emotions that are felt in response to those potential consequences. However, such a measure is not likely to capture the perceived probability of experiencing the outcomes, nor will it be as useful at understanding one's motivation to take mitigation action.     

Aging in the Era of Regenerative Medicine: Analysis of Aging‐Related Representations among Canadian Researchers

Explicitly aimed at understanding and controlling molecular and cellular processes at the root of senescence and biological aging, regenerative medicine aspires to artificially reproduce the biological processes that enable the body to regenerate itself. This no longer involves conserving the body's state of balance by combating disease, as in clinical medicine, but rather fighting degeneration itself. From stem cell research to gene therapy to the production of replacement tissues, regenerative medicine perfectly corresponds to the logic of biomedicalization specific to postmodern society. Based on a series of 18 interviews conducted with Canadian researchers and clinicians in the field of regenerative medicine, this article seeks to understand representations of the aging body among researchers in this field. Seen from a strictly negative angle, aging is assimilated by researchers to an inevitable catastrophe that nevertheless must be combated. More closely observing the theoretical model of regenerative biology and the types of treatments developed, it can be observed, however, that this medicine of the future does not target the elderly, but rather promises youth the ability to regenerate themselves to avoid aging.     

The intergenerational education spillovers of pension reform in China

Economic theory establishes that pension privatization weakens the link between old and young and so reduces the incentive to invest in public education in an economy with lower return rate of capital than growth rate of wage. However, empirical studies of the link change are few. In this paper, we investigate the effects of pension privatization and the central government’s subsidy to individual accounts on public education spending in a three-period overlapping generation model. And then, we take contemporary pension reforms in a number of Chinese provinces as offering natural experiment conditions. Using a difference-in-difference framework and 282 municipal districts panel data over years 1998–2009, we test the pension-education theoretical link change. Both our theoretical and empirical results confirm that pension privatization is adversely associated with local public spending on education in China. Private pension subsidies, moreover, magnify this effect. Our study supports the theoretical assertion and selective empirical findings of a negative intergenerational effect of pension privatization.     

Beyond tail median and conditional tail expectation: Extreme risk estimation using tail Lp-optimization

The conditional tail expectation (CTE) is an indicator of tail behavior that takes into account both the frequency and magnitude of a tail event. However, the asymptotic normality of its empirical estimator requires that the underlying distribution possess a finite variance; this can be a strong restriction in actuarial and financial applications. A valuable alternative is the median shortfall (MS), although it only gives information about the frequency of a tail event. We construct a class of tail L^p-medians encompassing the MS and CTE. For p in (1,2), a tail L^p-median depends on both the frequency and magnitude of tail events, and its empirical estimator is, within the range of the data, asymptotically normal under a condition weaker than a finite variance. We extrapolate this estimator and another technique to extreme levels using the heavy-tailed framework. The estimators are showcased on a simulation study and on real fire insurance data.