Evaluation of program success for programs with multiple trials in binary outcomes

A late‐stage clinical development program typically contains multiple trials. Conventionally, the program's success or failure may not be known until the completion of all trials. Nowadays, interim analyses are often used to allow evaluation for early success and/or futility for each individual study by calculating conditional power, predictive power and other indexes. It presents a good opportunity for us to estimate the probability of program success (POPS) for the entire clinical development earlier. The sponsor may abandon the program early if the estimated POPS is very low and therefore permit resource savings and reallocation to other products. We provide a method to calculate probability of success (POS) at an individual study level and also POPS for clinical programs with multiple trials in binary outcomes. Methods for calculating variation and confidence measures of POS and POPS and timing for interim analysis will be discussed and evaluated through simulations. We also illustrate our approaches on historical data retrospectively from a completed clinical program for depression. Copyright © 2015 John Wiley & Sons, Ltd.     

Notes on estimation of the intraclass correlation under the AB/BA crossover trial

Under the AB/BA crossover trial, we focus our attention on estimation of the intraclass correlation in normal data. We develop both point and interval estimators in closed form for the intraclass correlation. We employ Monte Carlo simulation to study the performance of these estimators in a variety of situations. We note that the estimators developed here for the intraclass correlation remain valid even when there are possibly unexpected carry-over effects.     

Construction of Simultaneous Confidence Intervals for Ratios of Means of Lognormal Distributions

For constructing simultaneous confidence intervals for ratios of means for lognormal distributions, two approaches using a two-step method of variance estimates recovery are proposed. The first approach proposes fiducial generalized confidence intervals (FGCIs) in the first step followed by the method of variance estimates recovery (MOVER) in the second step (FGCIs–MOVER). The second approach uses MOVER in the first and second steps (MOVER–MOVER). Performance of proposed approaches is compared with simultaneous fiducial generalized confidence intervals (SFGCIs). Monte Carlo simulation is used to evaluate the performance of these approaches in terms of coverage probability, average interval width, and time consumption.     

关于“概率论与数理统计”教学的几点新体会

提出改变传统的以教师为中心的教学结构,建立一种既能发挥教师的指导作用,又能充分体现学生主体作用的新型教学结构;采用中英文结合的教学模式,在以母语教学为主体的教学实践中贯穿双语教学,使学生能用英语进行数学方面的交流,并充分利用网络优势,扩大学生的知识面.     

Estimators for the Horvitz-Thompson Statistic Based on Some Posterior Distributions

The knowledge of first-order inclusion probabilities characterizing a sampling scheme is essential in design-based estimation of finite population totals. Sometimes the scheme is so complex that these probabilities cannot be computed exactly. Instead, both inclusion probabilities and corresponding sampling weights are simulated. One empirical Horvitz-Thompson estimator for a population total using simulation-based range-preserving estimates of sampling weights is obtained by applying the restricted maximum likelihood principle directly to each inclusion probability. The assumption of a prior distribution and the assessment of resulting posterior for a weight lead to two other estimators. One of them is the posterior mean estimator of the Horvitz-Thompson statistic. In a simulation involving Polish agricultural census data and a sequential fixed-cost sampling scheme, this estimator has attractive properties also from a frequentist point of view.     

Effects of Outcome and Probabilistic Ambiguity on Managerial Choices

Information ambiguity is prevalent in organizations and likely influences management decisions. This study examines, given imprecise probabilities and outcomes, how managers make choices when they are provided with single-figure benchmarks. Seventy-nine MBA students completed two experiments. We found that, in a decision framed as a decision under certainty involving an ambiguous outcome, the majority of the subjects were ambiguity prone in the loss condition and switched to ambiguity aversion in the gain condition. However, in the presence of probabilistic ambiguity in a decision under risk, this expected switching pattern was shown only when the difference in riskiness between the two choice options (in the loss condition) was perceived to be relatively small. In a companion study, we used a written protocol approach to identify factors that affect decision makers' investment choices when faced with ambiguous outcomes. Protocols frequently mentioned that the ambiguous outcome option was risky, even in the case which was framed as a decision under certainty in the problem statement. In a decision under risk with ambiguous outcomes, the combination of probabilistic risk and outcome ambiguity was seen as even more risky.     

Consequentialist foundations for expected utility

Behaviour norms are considered for decision trees which allow both objective probabilities and uncertain states of the world with unknown probabilities. Terminal nodes have consequences in a given domain. Behaviour is required to be consistent in subtrees. Consequentialist behaviour, by definition, reveals a consequence choice function independent of the structure of the decision tree. It implies that behaviour reveals a revealed preference ordering satisfying both the independence axiom and a novel form of sure-thing principle. Continuous consequentialist behaviour must be expected utility maximizing. Other plausible assumptions then imply additive utilities, subjective probabilities, and Bayes' rule.     

On the emptiness of the stability set of order d

We know from Li's theorem (1993) that the stability set of order d may be empty for some preference profiles. However, one may wonder whether such situations are just rare oddities or not. In this paper, we partially answer this question by considering the restrictive case where the number of alternatives is the smallest compatible with an empty stability set. More precisely, we provide an upper bound on the probability for having an empty stability set of order d for the majority game under the Impartial Weak Ordering Culture assumption. This upper bound is already extremely low for small population and tends to zero as the number of individuals goes to infinity.     

Paying to improve your chances: Gambling or insurance?

Will a more risk-averse individual spend more or less to improve probabilities, say on marketing efforts that enhance the chance of a sale? For any two payoffs and starting probabilities, the answer is unfortunately indeterminate. However, interpreting gambling as increasing small chances of good outcomes and insurance as reducing small chances of bad outcomes, the more risk-averse individual will pay less (more) to gamble (insure). We find a critical switching probability that depends on the individuals and outcomes involved. If the good outcome is less (more) likely than this critical value, the expenditures represent gambling (insurance).     

Buffon's needle and the estimation of migration distances

A procedure is suggested for estimating migration distances from data on the proportion of migrants crossing regional boundaries. The method makes use of Buffon's needle, a problem in geometrical probability from the eighteenth century that was originally used as an empirical means of estimating π. The procedure is described for various scenarios that differ in their assumptions about region shape, the spatial distribution of population, and the distribution of migration distances. An application to migration distances in the United States is given, and additional attention is given to the estimation of intraregional migration distances.