Estimation of cardinal utility based on a nonlinear theory

Empirical studies have demonstrated that cardinal utility functions assessed via gamble-based methods are often incoherent because of the probability and certainty effects. These effects are caused by apparent risk attitudes different from those admissible within the linear (expected) utility theory. The incoherences can also be accentuated by the effects of chaining and serial positioning of responses. To filter out these effects and obtain an unbiased measurement of the strength of preference, and a simultaneous measurement of risk attitude, we devised the independent-gamble, nonlinear-inference (IGNI) method: the utility function of outcomes and the risk function of probabilities are estimated jointly from assessed certainty equivalents of independent gambles by using a nonlinear utility theory for inference. The method contrasts with all popular utility assessment techniques in that it estimates a cardinal function in the two-dimensional space of outcomes and probabilities. Hence, it allows us to obtain novel insights into the nature of utility functions and the probability effect. Both are illustrated by empirical results for fifty-four subjects.     

On Random Sampling Without Replacement from a Finite Population

We consider the three progressively more general sampling schemes without replacement from a finite population: simple random sampling without replacement, Midzuno sampling and successive sampling. We (i) obtain a lower bound on the expected sample coverage of a successive sample, (ii) show that the vector of first order inclusion probabilities divided by the sample size is majorized by the vector of selection probabilities of a successive sample, and (iii) partially order the vectors of first order inclusion probabilities for the three sampling schemes by majorization. We also show that the probability of an ordered successive sample enjoys the arrangement increasing property and for sample size two the expected sample coverage of a successive sample is Schur convex in its selection probabilities. We also study the spacings of a simple random sample from a linearly ordered finite population and characterize in several ways a simple random sample.     

EFFECTS OF INDIVIDUAL HETEROGENEITY IN ESTIMATING THE PERSISTENCE OF SMALL POPULATIONS

ABSTRACT. Population viability models are commonly used to estimate the probability of persistence of small, threatened, or endangered populations. Demographic, temporal, spatial, and individual heterogeneity are important factors affecting the probability of persistence of small populations. Because stochastic process are intractable analytically (Lud-wig [1996]), computer simulation models are often used for estimating population viability via numerical techniques. Although demographic, spatial, and temporal stochasticity have been incorporated into some population viability models, individual heterogeneity has not been included. In this paper we include individual heterogeneity in a simulation model and examine probabilities of population persistence at different levels of heterogeneity and population size. Individual heterogeneity may increase the probability of persistence of small populations. The mechanism for the extension in persistence may be explained by natural selection. Genotypes persisting through a decline may be those that survive better under the conditions causing the decline. These individuals that survive and reproduce in the face of adverse conditions may extend the probability that a small population persists.     

Risk processes with shot noise Cox claim number process and reserve dependent premium rate

We consider a suitable scaling, called the slow Markov walk limit, for a risk process with shot noise Cox claim number process and reserve dependent premium rate. We provide large deviation estimates for the ruin probability. Furthermore, we find an asymptotically efficient law for the simulation of the ruin probability using importance sampling. Finally, we present asymptotic bounds for ruin probabilities in the Bayesian setting.     

Estimating tail probabilities of heavy tailed distributions with asymptotically zero relative error

Efficient estimation of tail probabilities involving heavy tailed random variables is amongst the most challenging problems in Monte-Carlo simulation. In the last few years, applied probabilists have achieved considerable success in developing efficient algorithms for some such simple but fundamental tail probabilities. Usually, unbiased importance sampling estimators of such tail probabilities are developed and it is proved that these estimators are asymptotically efficient or even possess the desirable bounded relative error property. In this paper, as an illustration, we consider a simple tail probability involving geometric sums of heavy tailed random variables. This is useful in estimating the probability of large delays in M/G/1 queues. In this setting we develop an unbiased estimator whose relative error decreases to zero asymptotically. The key idea is to decompose the probability of interest into a known dominant component and an unknown small component. Simulation then focuses on estimating the latter ‘residual’ probability. Here we show that the existing conditioning methods or importance sampling methods are not effective in estimating the residual probability while an appropriate combination of the two estimates it with bounded relative error. As a further illustration of the proposed ideas, we apply them to develop an estimator for the probability of large delays in stochastic activity networks that has an asymptotically zero relative error.      

On linear classification procedures between two categories with known mean vectors and covariance matrices

Summary This paper is concerned with probabilities (error probabilities), caused by misclassification, of linear classification procedures (linear procedures) between two categories, whose mean vectors and covariance matrices are assumed to be known, while the distribution of each category may well be continuous or discrete. The tightest upper bounds on the largest of two kinds of error probability of each linear procedure and on the expected error probability for any apriori probabilities are obtained. Moreover in some cases of interest, theoptimal linear procedure (in the sense of attaining the infimum out of all the upper bounds) is given.     

Multidimensional occupancy problems with Poisson randomization

In this paper the concept of Poisson randomization is studied as given in [1, 2] and analogous formulae for the generalized process are derived. The generalization regards occupancy problems where different ball types are considered such that each type has an associated probability distribution of urn occupancy. Theorems are given for formulae to calculate probabilities of events and the distribution and moments of waiting time random variables. Finally, the theory is illustrated with examples.     

Markovian Imperfect Software Debugging Model and Its Performance Measures

Abstract

This paper considers a Markovian imperfect software debugging model incorporating two types of faults and derives several measures including the first passage time distribution. When a debugging process upon each failure is completed, the fault which causes the failure is either removed from the fault contents with probability p or is remained in the system with probability 1 ? p. By defining the transition probabilities for the debugging process, we derive the distribution of first passage time to a prespecified number of fault removals and evaluate the expected numbers of perfect debuggings and debugging completions up to a specified time. The availability function of a software system, which is the probability that the software is in working state at a given time, is also derived and thus, the availability and working probability of the software system are obtained. Throughout the paper, the length of debugging time is treated to be random and thus its distribution is assumed. Numerical examples are provided for illustrative purposes.     

A Unified Derivation of the Complementary Waiting Time Distribution in Sequential Occupancy

Occupancy distributions are defined on the stochastic model of random allocation of balls to a specific number of distinguishable urns. The reduction of the joint distribution of the occupancy numbers, when a specific number of balls are allocated, to the joint conditional distribution of independent random variables given their sum, when the number of balls allocated is unspecified, is a powerful technique in the study of occupancy distributions. Consider a supply of balls randomly distributed into n distinguishable urns and assume that the number X of balls distributed into any specific urn is a random variable with probability function P(X = x) = q _x , x = 0, 1,.... The probability function of the number L _r of occupied urns until r balls are placed into previously occupied urns is derived in terms of convolutions of q _x , x = 0, 1,... and their finite differences. Further, using this distribution, the minimum variance unbiased estimator of the parameter n, based on a suitable sequential sampling scheme, is deduced. Finally, some illustrating applications are discussed.      

Sampling scattered data with Bernstein polynomials: stochastic and deterministic error estimates

Viewing the classical Bernstein polynomials as sampling operators, we study a generalization by allowing the sampling operation to take place at scattered sites. We utilize both stochastic and deterministic approaches. On the stochastic side, we consider the sampling sites as random variables that obey some naturally derived probabilistic distributions, and obtain Chebyshev type estimates. On the deterministic side, we incorporate the theory of uniform distribution of point sets (within the framework of Weyl’s criterion) and the discrepancy method. We establish convergence results and error estimates under practical assumptions on the distribution of the sampling sites.     

KAPPA FUNCTION AS A UNIFYING FRAMEWORK FOR DISCRETE POPULATION MODELING

This paper develops a unified way to describe the various generalized discrete‐time nonlinear dynamical models with density dependence, Allee effects, and parasitoids. We show how the kappa function can be used to describe the probabilities involved in intra‐ or interspecific encounters, namely, (i) the probability of surviving to the next generation in the absence of parasitoids or Allee effects, (ii) the encounter probability associated with Allee effects, and (iii) the probability of escaping parasitism in the presence of parasitoids. Having introduced a phenomenological framework of modeling via the kappa function, we then provide a realistic mechanism through stochastic encounters, responsible for generating the kappa function to any of the three involved probabilities. The unified modeling through the kappa function yields insights into how abundances influence species interactions. It is now straightforward to use this unified modeling to analyze and investigate its consequences in species dynamics.     

Estimating completion-time distribution in stochastic activity networks

This paper deals with simulation-based estimation of the probability distribution for completion time in stochastic activity networks. These distribution functions may be valuable in many applications. A simulation method, using importance-sampling techniques, is presented for estimation of the probability distribution function. Separating the state space into two sets, one which must be sampled and another which need not be, is suggested. The sampling plan of the simulation can then be decided after the probabilities of the two sets are adjusted. A formula for the adjustment of the probabilities is presented. It is demonstrated that the estimator is unbiased and the upper bound of variance minimized. Adaptive sampling, utilizing the importance sampling techniques, is discussed to solve problems where there is no information or more than one way to separate the state space. Examples are used to illustrate the sampling plan.     

Effects of heterogeneity on bank efficiency scores

Bank efficiency estimates often serve as a proxy of managerial skill since they quantify sub-optimal production choices. But such deviations can also be due to omitted systematic differences among banks. In this study, we examine the effects of heterogeneity on bank efficiency scores. We compare different specifications of a stochastic cost and alternative profit frontier model with a baseline specification. After conducting a specification test, we discuss heterogeneity effects on efficiency levels, ranks and the tails of the efficiency distribution. We find that heterogeneity controls influence both banks’ optimal costs and profits and their ability to be efficient. Differences in efficiency scores are important for more than only methodological reasons. First, different ways of accounting for heterogeneity result in estimates of foregone profits and additional costs that are significantly different from what we infer from our general specification. Second, banks are significantly re-ranked when their efficiency is estimated with a specification other than the preferred, general specification. Third, the general specification gives the most reliable estimates of the probability of distress, although differences to the other specifications are low.     

Slowdown and neutral pockets for a random walk in random environment

We consider a d-dimensional random walk in random environment for which transition probabilities at each site are either neutral or present an effective drift “pointing to the right”. We obtain large deviation estimates on the probability that the walk moves in a too slow ballistic fashion, both under the annealed and quenched measures. These estimates underline the key role of large neutral pockets of the medium in the occurrence of slowdowns of the walk. Received: 12 March 1998 / Revised version: 19 February 1999     

HABITAT PATCH OCCUPANCY DYNAMICS OF GLAUCOUS‐WINGED GULLS (LARUS GLAUCESCENS) I: A DISCRETE‐TIME MODEL

ABSTRACT. Diurnal habitat occupancy dynamics of Glaucous‐winged Gulls were evaluated in a system of six habitats on and around Protection Island, Washington. Data were collected on the rates of gull movement between habitat patches, and from these data the probabilities of transitions between habitats were estimated as functions of tide height and time of day. A discrete‐time matrix model based on the transition probabilities was used to generate habitat occupancy predictions, which were then compared to hourly census data. All model parameters were estimated directly from data rather than through model fitting. The model made reasonable predictions for two of the six habitats and explained 45% of the variability in the data from 2003. The construction and testing of mathematical models that predict occupancies in multiple habitats may play increasingly important roles in the understanding and management of animal populations within complex environments.     

Predicting Web Users' Next Access Based on Log Data

This article considers models that describe how people browse the Web. We restrict our attention to navigation patterns within a single site, and base our study on standard Web server access logs. Given a visitor's previous activities on the site, we propose models that predict their next page request. If the prediction is reasonably accurate, we might consider “prefetching” the page before the visitor requests it. A more conservative use for such predictions would be to simply update the freshness records in a proxy or network cache, eliminating unnecessary If-Modified-Since requests. Using data from the Web site for the Computing and Mathematical Sciences Research Division of Lucent Technologies (cm.bell-labs.com) we first evaluate the predictive performance of low-order Markov models. We next consider mixtures of first-order Markov models, achieving a kind of clustering of Web pages in the site. This approach is shown to perform well, while significantly reducing the space required to store the model. Finally, we explore a Bayesian approach using a Dirichlet prior on the collection of links available to a user at each stage in their travels through the site. We show that the posterior probabilities derived under this model are fairly close to the cross-validation estimates of the probability of success.     

THE EFFECTS OF REDUCING MATING LIKELIHOOD ON POPULATION VIABILITY

Abstract The success a species may have invading a patch previously unoccupied is of considerable interest for pest managers and conservation ecologists. The purpose here is to present a mechanistic approach to analyze reproductive Allee effects appearing through the failure in the process of fertilization in a two‐sex population and observe how the survival in an invaded patch is affected. This is in contrast to the usually employed stochastic models with a deterministic skeleton that describe the presence of Allee effects. A Poisson–Ricker model, which includes stochastic demography and sex determination with females classified as successfully fertilized or not fertilized, is used. Numerical approximations to the probabilities of extinction and the mean time to extinction are presented, for fixed parameter values, suggesting how stochasticity in the mating process combined with random fluctuations in the male and female densities, at each generation, contribute to the risk of extinction of a population which started an invasion at a low density.     

Survival risks,intertemporal consumption,and insurance: The case of distorted probabilities

This paper explores how to evaluate changes in survival probabilities when people do not process probabilities linearly, as is commonly assumed in the literature, but distort probabilities. We show that the valuation of risks to life depends critically on two parameters: the elasticity of the probability weighting function and the elasticity of the utility function with respect to future consumption. Using estimates from the empirical literature we derive that the bias of erroneously ignoring probability distortion in general leads to cost–benefit ratios that are too high and that generate too much priority for programs that save young lives.     

The Construction and Properties of Boundary Kernels for Smoothing Sparse Multinomials

In recent years several authors have investigated the use of smoothing methods for sparse multinomial data. In particular, Hall and Titterington (1987) studied kernel smoothing in detail. It is pointed out here that the bias of kernel estimates of probabilities for cells near the boundaries of the multinomial vector can dominate the mean sum of squared error of the estimator for most true probability vectors. Fortunately, boundary kernels devised to correct boundary effects for kernel regression estimators can achieve the same result for these estimators. Properties of estimates based on boundary kernels are investigated and compared to unmodified kernel estimates and maximum penalized likelihood estimates. Monte Carlo evidence indicates that the boundary-corrected kernel estimates usually outperform uncorrected kernel estimates and are quite competitive with penalized likelihood estimates.     

Never say “not”: Impact of negative wording in probability phrases on imprecise probability judgments

Effective translations between numerical and verbal representations of uncertainty are a concern shared by researchers in cognitive science and psychology, with applications to real-world risk management and decision support systems. While there is a substantial literature on such translations for point-wise probabilities, this paper contributes to the scanty literature on imprecise probability translations. Reanalysis of Budescu et al.’s [1] data on numerical interpretations of the Intergovernmental Panel on Climate Change [2] fourth report’s verbal probability expressions (PEs) revealed that negative wording has deleterious effects on lay judgements. Budescu et al. asked participants to interpret PEs in IPCC report sentences, by asking them to provide lower, “best” and upper estimates of the probabilities that they thought the authors intended. There were four experimental conditions, determining whether participants were given any numerical guidelines for translating the PEs into numbers.The first analysis focuses on twelve sentences in Budescu et al. that used the PE “very likely,” “likely,” “unlikely,” or “very unlikely”. A mixed beta regression modelling the lower, “best” and upper estimates revealed a less regressive mean and less dispersion for positive than for negative wording in all three estimates, for both the “very likely” and “likely” sentence sets. The Budescu et al. data also included a task asking for context-free translations of these PEs, and a similar pattern of results was found for that task. Negative wording therefore resulted in more regressive estimates and less consensus regardless of experimental condition.The second analysis focuses on two statements that were positive–negative duals. Appropriate pairs of responses were assessed for conjugacy and additivity. A large majority of respondents were appropriately super- and sub-additive in their lower and upper probability estimates. A mixed beta regression model of these three variables revealed that respondents were suprisingly close to obeying the conjugacy relationships for lower and upper probabilities.