Models and Inference for Corrosion Pit Depth Data

A new regression structure is introduced to several families of distributions, including the Generalized Extreme Value (GEV) distribution, which are available to explain the distribution of the maximum pit depth of a small sample. This regression structure depends on sample sizes, but under a simple assumption it is also available in the case of unknown sample sizes. GEV with the proposed regression structure provides us with a new extrapolation model for time. Compared to ordinary GEV models, effectiveness of our GEV model is demonstrated for our real data set. Also, comparisons are made with GEV and other two families of distributions with respect to the proposed regression structure.     

A theoretical analysis of the cross-nested logit model

The emergence of Intelligent Transportation Systems and the associated technologies has increased the need for complex models and algorithms. Namely, real-time information systems, directly influencing transportation demand, must be supported by detailed behavioral models capturing travel and driving decisions. Discrete choice models methodology provide an appropriate framework to capture such behavior. Recently, the Cross-Nested Logit (CNL) model has received quite a bit of attention in the literature to capture decisions such as mode choice, departure time choice and route choice. %The CNL model is an extension of the Nested Logit model, providing %more flexibility at the cost of some complexity in the model formulation. In this paper, we develop on the general formulation of the Cross Nested Logit model proposed by Ben-Akiva and Bierlaire (1999) and based on the Generalized Extreme Value (GEV) model. We show that it is equivalent to the formulations byby Papola (2004) and Wen and Koppelman (2001). We also show that the formulations by Small(1987) and Vovsha(1997) are special cases of this formulation. We formally prove that the Cross-Nested Logit model is indeed a member of the GEV models family. In doing so, we clearly distinguish between conditions that are necessary to prove consistency with the GEV theory, from normalization conditions. Finally, we propose to estimate the model with non-linear programming algorithms, instead of heuristics proposed in the literature. In order to make it operational, we provide the first derivatives of the log-likelihood function, which are necessary to such optimization procedures.     

Synoptic airflow and UK daily precipitation extremes

We develop a vector generalised linear model to describe the influence of the atmospheric circulation on extreme daily precipitation across the UK. The atmospheric circulation is represented by three covariates, namely synoptic scale airflow strength, direction and vorticity; the extremes are represented by the monthly maxima of daily precipitation, modelled by the generalised extreme value distribution (GEV). The model parameters for data from 689 rain gauges across the UK are estimated using a maximum likelihood estimator. Within the framework of vector generalised linear models, various plausible models exist to describe the influence of the individual covariates, possible nonlinearities in the covariates and seasonality. We selected the final model based on the Akaike information criterion (AIC), and evaluated the predictive power of individual covariates by means of quantile verification scores and leave-one-out cross validation. The final model conditions the location and scale parameter of the GEV on all three covariates; the shape parameter is modelled as a constant. The relationships between strength and vorticity on the one hand, and the GEV location and scale parameters on the other hand are modelled as natural cubic splines with two degrees of freedom. The influence of direction is parameterised as a sine with amplitude and phase. The final model has a common parameterisation for the whole year. Seasonality is partly captured by the covariates themselves, but mostly by an additional annual cycle that is parameterised as a phase-shifted sine and accounts for physical influences that we have not attempted to explicitly model, such as humidity.     

Numerical scaling of ratio scale utilities in multi-criteria decision analysis with geometric model

The ratio scale measurement of utilities in multi-criteria decision analysis has been criticised due to theoretical and practical interpretation of the required assessments. Compared to the ratio scale models, for example, the ‘0–1-scale’ models based on interval scale have been more generally accepted. On the other hand, under some conditions it has been shown that the results of multi-criteria decision support based on ratio scale comparisons are actually independent of the numerical magnitude of the ratio scale assessments. This paper reviews and interprets these results and proposes new solutions for the remaining problems, for which the numerical magnitude of the ratio scale utilities is required. Moreover, we interpret the scale-independent ratio scale models with respect to some other multi-criteria decision support models. The theoretical analysis is carried out by statistical models and illustrated by numerical examples.     

Sources of uncertainty in the extreme value statistics of climate data

We investigate three sources of uncertainty in the calculation of extreme value statistics for observed and modeled climate data. Inter-model differences in formulation, unforced internal variability and choice of statistical model all contribute to uncertainty. Using fits to the GEV distribution to obtain 20 year return values, we quantify these uncertainties for the annual maximum daily mean surface air temperatures of pre-industrial control runs from 15 climate models in the CMIP3 dataset.     

Extreme value distributions of inclusions in six steels

There is a prevailing assumption that the largest inclusions in steel volumes follows mode I of the Generalized Extreme Values (GEV) distribution. In this work, the GEV distributions of non-metallic inclusions in six different high performance steels, of different grades and processing routes, were investigated by means of fractography of inclusions causing failure in ultrasonic fatigue testing to one billion cycles and all three modes of the GEV were found for the different steel grades. Values of the shape parameter ξ of the GEV distribution as high as 0.51 (standard deviation 0.11) were found in one steel grade. Thus, the present results show that the assumption of GEV-I (Gumbel, LEVD) distribution has to be substantiated before being used to estimate the size of the largest inclusions.     

Nonlinear functional analysis and optimal economic growth

A problem of existence and characterization of solutions of optimal growth models in many sector economies is studied. The social utility to be optimized is a generalized form of a preference depending additively on consumption at the different dates of the planning period. The optimization is restricted to a set of admissible growth paths defined by production-investment-consumption relations described by a system of differential equations. Sufficient conditions are given for existence of a solution in a Hilbert space of paths, without convexity assumptions on either the utilities or the technology, using techniques of nonlinear functional analysis. A characterization is given of the utilities which are continuous with respect to the Hilbert space norm. Under convexity assumptions a characterization is also given of optimal and efficient solutions by competitive prices.     

Oligopolistic markets with leadership,cooperative followers and a multivalued inverse demand function

We consider an oligopolistic market with a leader, finitely many followers and a multivalued inverse demand function. By assuming that there exist cooperative behaviors of followers, we introduce the notions of Stackelberg Cournot equilibria for markets with nontransferable utilities and transferable utilities. We shall prove their existence theorems in our models.     

Practical Extreme Value Modelling of Hydrological Floods and Droughts: A Case Study

Estimation of flood and drought frequencies is important for reservoir design and management, river pollution, ecology and drinking water supply. Through an example based on daily streamflow observations, we introduce a stepwise procedure for estimating quantiles of the hydrological extremes floods and droughts. We fit the generalised extreme value (GEV) distribution by the method of block maxima and the generalised Pareto (GP) distribution by applying the peak over threshold method. Maximum likelihood, penalized maximum likelihood and probability weighted moments are used for parameter estimation. We incorporate trends and seasonal variation in the models instead of splitting the data, and investigate how the observed number of extreme events, the chosen statistical model, and the parameter estimation method effect parameter estimates and quantiles. We find that a seasonal variation should be included in the GEV distribution fitting for floods using block sizes less than one year. When modelling droughts, block sizes of one year or less are not recommended as significant model bias becomes visible. We conclude that the different characteristics of floods and droughts influence the choices made in the extreme value modelling within a common inferential strategy.This revised version was published online in March 2005 with corrections to the cover date.     

Ordinal utility and economic price indices

Summary In economic price index theory, a reference level of utility is needed for measuring the change in the cost of living oetween a base period and a comparison period. A reference level function can be used to derive this reference utility level from the utilities attained at the base and at the comparison prices. Depending on the scale type of the underlying utility function, the reference level function has to satisfy certain invariance conditions. In this paper, these conditions are formulated as functional equations for interval scales and for ordinal utility scales. By solving these equations, we characterize the class of admissible reference level functions for the respective scale type.We thank J. Aczél and an anonymous referee for their comments on an earlier version of the paper.     

Preference relations on a set of fuzzy utilities as a basis for decision making

In fuzzy decision problems, we often encounter situations of choosing among alternatives which are assigned fuzzy utilities. These problems have been approached using fuzzy implications or direct comparisons among fuzzy utilities. In the literature, however, there are few attempts to investigate the issues addressing reasonable choice or reasonable ordering using fuzzy sets theory. This paper first introduces some fundamental properties of fuzzy binary relations and certain conditions of reasonable orderings of fuzzy utilities. Then a method for constructing a fuzzy preference relation on a given set of fuzzy utilities is proposed for the sake of rational decision making. This procedure employs the concepts of the extended minimum and the Hamming distance between the greatest upper sets or the greatest lower sets of fuzzy utilities. Finally it is shown that the proposed fuzzy preference relations have reasonable properties as fuzzy orderings for decision making.     

On the maximum likelihood estimator for the Generalized Extreme-Value distribution

The vanilla method in univariate extreme-value theory consists of fitting the three-parameter Generalized Extreme-Value (GEV) distribution to a sample of block maxima. Despite claims to the contrary, the asymptotic normality of the maximum likelihood estimator has never been established. In this paper, a formal proof is given using a general result on the maximum likelihood estimator for parametric families that are differentiable in quadratic mean but whose supports depend on the parameter. An interesting side result concerns the (lack of) differentiability in quadratic mean of the GEV family.     

Detecting change in UK extreme precipitation using results from the climate<Emphasis Type="Italic">prediction</Emphasis>.net BBC climate change experiment

We investigate a question posed by policy makers, namely, “when will changes in extreme precipitation due to climate change be detectable?” To answer this question we use climateprediction.net (CPDN) model simulations from the BBC Climate Change Experiment (CCE) over the UK. These provide us with the unique opportunity to compare 1-day extreme precipitation generated from climate altered by observed forcings (time period 1920–2000) and the SRES A1B emissions scenario (time period 2000–2080) (the Scenario) to extreme precipitation generated by a constant climate for year 1920 (the Control) for the HadCM3L General Circulation Model (GCM). We fit non-stationary Generalized Extreme Value (GEV) models to the Scenario output and compare these to stationary GEV models fit to the parallel Control. We define the time of detectable change as the time at which we would reject a hypothesis at the α = 0.05 significance level that the 20-year return level of the two runs is equal. We find that the time of detectable change depends on the season, with most model runs indicating that change to winter extreme precipitation may be detectable by the year 2010, and that change to summer extreme precipitation is not detectable by 2080. We also investigate which climate model parameters affect the weight of the tail of the precipitation distribution and which affect the time of detectable change for the winter season. We find that two climate model parameters have an important effect on the tail weight, and two others seem to affect the time of detection. Importantly, we find that climate model simulated extreme precipitation has a fundamentally different behavior to observations, perhaps due to the negative estimate of the GEV shape parameter, unlike observations which produce a slightly positive (～0.0–0.2) estimate.     

Supply chain network sustainability under competition and frequencies of activities from production to distribution

In this paper, we develop a competitive supply chain network model with multiple firms, each of which produces a differentiated product by brand and weights the emissions that it generates through its supply chain network activities in an individual way. The supply chain network activities of production, transport and distribution, and storage have associated with them distinct capacities and the firms seek to determine their optimal product flows and frequencies of operation so that their utilities are maximized where the utilities consist of profits and weighted emissions. Multiple production, storage, and transport mode options are allowed. The governing equilibrium concept is that of Cournot–Nash equilibrium. We provide both path and link flow variational inequality formulations of the equilibrium conditions and then propose an algorithm, which, at each iteration, yields closed form expressions for the underlying variables. Numerical examples illustrate the generality of the model and the information provided to managerial decision-makers and policy-makers. This paper adds to the growing literature on sustainable supply chains through the development of a computable general competitive supply chain network game theory model, which brings a greater realism to the evaluation of profit and emission trade-offs through the incorporation of frequencies.     

Forward dynamic utility functions: A new model and new results

A major obstacle in the existing models of forward dynamic utilities and investment performance evaluation is to establish the existence and uniqueness of the optimal solutions. Consequently, we present a new model of forward dynamic utilities. In doing so, we establish the existence and uniqueness of the solutions for a general (smooth) utility function, and we show that the assumptions needed for such solutions are similar to those under the backward formulation. Moreover, we provide unique viscosity solutions. We also provide discontinuous viscosity solutions. In addition, we introduce Hausdorff-continuous viscosity solutions to the portfolio model.     

跨境水资源竞争制度模型与竞争行为

全球淡水资源稀缺与分布不平衡导致跨境水资源竞争及水安全问题日趋严峻。为了抑制过度竞争、提高流域整体效用、促进流域合作和引导生态环保行为,提出跨境水资源竞争的三种改进制度。运用制度结构图与数学模型,比较不同制度结构下流域整体效用和各国竞争行为,验证了改进制度的有效性。据此,提出我国参与跨境水资源竞争和管控的策略建议。     

On variable discounting in dynamic programming: applications to resource extraction and other economic models

This paper generalizes the classical discounted utility model introduced in Samuelson (Rev. Econ. Stud. 4:155–161, 1937) by replacing a constant discount rate with a function. The existence of recursive utilities and their constructions are based on Matkowski’s extension of the Banach Contraction Principle. The derived utilities go beyond the class of recursive utilities studied in the literature and enable a discussion on subtle issues concerning time preferences in the theory of finance, economics or psychology. Moreover, our main results are applied to the theory of optimal economic growth related with resource extraction models with unbounded utility function of consumption.     

A Framework Algorithm to Compute Optimal Asset Allocation for Retirement with Behavioral Utilities

The question of optimal strategic asset allocation for investors with behavioural utilities saving for retirement is addressed. To date this problem has been studied assuming that an investor is rational in the sense when making investment decisions the preference relation of the investor satisfies all the axioms of choice. Research in behavioural science indicates that investment related decisions of many people do not satisfy the axioms of choice. Our interest is in developing a platform that allows the use of a broader class of utilities that may or may not satisfy the axioms of choice. Such utilities may not be convex. Our interest is in developing a framework algorithm that enables a user considerable flexibility in how their needs may be specified. For illustrative purposes a binomial tree is used to model asset returns, although the method developed can be used with more elaborate models.Work supported by the National Science Foundation grant CCR-9988205 and Office of Naval Research grant N00014-96-1-0274.     

Combinatorial criteria for uniqueness of Gibbs measures

We generalize previously known conditions for uniqueness of the Gibbs measure in statistical physics models by presenting conditions of any finite size for models on any underlying graph. We give two dual conditions, one requiring that the total influence on a site is small, and the other that the total influence of a site is small. Our proofs are combinatorial in nature and use tools from the analysis of discrete Markov chains, in particular the path coupling method. The implications of our conditions for the mixing time of natural Markov chains associated with the models are discussed as well. We also present some examples of models for which the conditions hold. © 2005 Wiley Periodicals, Inc. Random Struct. Alg., 2005     

Optimal contracts in portfolio delegation

The optimal contracts in portfolio delegation under general preferences are characterized when the underlying state variable is not contractible, and the principal must rely on the final returns of portfolios to design the compensation schemes for the fund manager. We show that the optimal contracts satisfy a second-order nonlinear ordinary differential equation that depends on the utility functions and the distribution of state price density. In general, there is an efficiency loss for the optimal contracts unless the utility functions of both the principal and the agent exhibit linear risk tolerance with identical cautiousness. Additional contractible observables, like stock indexes, can be used to improve the efficiency of the second-best contracts, even if they are not perfectly correlated with the underlying state price. A continuous-time example with power utilities is presented to illustrate the features of the optimal contracts.