A STATE SPACE BIOECONOMIC MODEL OF PACIFIC HALIBUT

Abstract Evaluation of potential economic consequences of alternative management actions requires an understanding of how the biological stock will be affected by the management action and an understanding of the response of economic systems to changes in the timing, magnitude, and size distribution of harvests and changes in the location and catchability of the biological stock. We use a hybrid structural time series model to represent Pacific halibut (Hippoglossus stenolepis) stock and recruitment dynamics and a system of structural equations to represent supply and demand relationships for Pacific halibut from Alaska and British Columbia. Model simulations explore the economic effects of changes in recruitment success, growth rate, and carrying capacity, and changes in international supplies of halibut.     

Universality of Chaos and Ultrametricity in Mixed p‐Spin Models

We prove disorder universality of chaos phenomena and ultrametricity in the mixed p‐spin model under mild moment assumptions on the environment. This establishes the longstanding belief among physicists that the solution of mean‐field models with Gaussian disorder also holds for different environments. Our results extend to the mixed p‐spin model as well as to different spin glass models. These include universality of quenched disorder chaos in the Edwards‐Anderson (EA) model and quenched concentration for the magnetization in both EA and mixed p‐spin models under non‐Gaussian environments. In addition, we show quenched self‐averaging for the overlap in the random field Ising model under small perturbation of the external field.© 2015 Wiley Periodicals, Inc.     

MODELING INDIVIDUAL TREE MORTALITY RATES USING MARGINAL AND RANDOM EFFECTS REGRESSION MODELS

Abstract Developing models to predict tree mortality using data from long‐term repeated measurement data sets can be difficult and challenging due to the nature of mortality as well as the effects of dependence on observations. Marginal (population‐averaged) generalized estimating equations (GEE) and random effects (subject‐specific) models offer two possible ways to overcome these effects. For this study, standard logistic, marginal logistic based on the GEE approach, and random logistic regression models were fitted and compared. In addition, four model evaluation statistics were calculated by means of K‐fold cross‐valuation. They include the mean prediction error, the mean absolute prediction error, the variance of prediction error, and the mean square error. Results from this study suggest that the random effects model produced the smallest evaluation statistics among the three models. Although marginal logistic regression accommodated for correlations between observations, it did not provide noticeable improvements of model performance compared to the standard logistic regression model that assumed impendence. This study indicates that the random effects model was able to increase the overall accuracy of mortality modeling. Moreover, it was able to ascertain correlation derived from the hierarchal data structure as well as serial correlation generated through repeated measurements.     

Estimation of integrated volatility in stochastic volatility models

In the framework of stochastic volatility models we examine estimators for the integrated volatility based on the pth power variation (i.e. the sum of pth absolute powers of the log‐returns). We derive consistency and distributional results for the estimators given high‐frequency data, especially taking into account what kind of process we may add to our model without affecting the estimate of the integrated volatility. This may on the one hand be interpreted as a possible flexibility in modelling, for example adding jumps or even leaving the framework of semimartingales by adding a fractional Brownian motion, or on the other hand as robustness against model misspecification. We will discuss possible choices of p under different model assumptions and irregularly spaced data. Copyright © 2005 John Wiley & Sons, Ltd.     

MULTISPECIES AND SINGLE‐SPECIES MODELS OF FISH POPULATION DYNAMICS: COMPARING PARAMETER ESTIMATES

Abstract Population features inferred from single‐species, age‐structured models are compared to those inferred from a multispecies, age‐structured model that includes predator‐prey interactions among three commercially harvested fish species—walleye pollock, Atka mackerel, and Pacific cod—on the Aleutian Shelf, Alaska. The multispecies framework treats the single‐species models and data as a special case of the multispecies model and data. The same data from fisheries and surveys are used to estimate model parameters for both single‐species and multispecies configurations of the model. Additionally, data from stomach samples and predator rations are used to estimate the parameters of the multispecies model. One form of the feeding functional response, predator pre‐emption, was selected using AIC from seven alternative models for how the predation rate changes with the densities of prey and possibly other predators. Differences in estimated population dynamics and productivity between the multispecies and single‐species models were observed. The multispecies model estimated lower mackerel population sizes from 1964–2003 than the single‐species model, while the spawning biomass of pollock was estimated to have declined more than three times faster since 1964 by the multispecies model. The variances around the estimates of spawning biomass were smaller for mackerel and larger for pollock in the multispecies model compared to the single‐species model.     

Wavelet Analysis and Chirps

In this article, we propose a general model for some chirp-like signals. The analytical structure of these chirps is given in the form |x|^sg(x/|x|^1+β), wheregis aL^pindefinitely oscillating function like sin(x). Under suitable assumptions, we achieve a characterization of the chirps by means of their wavelet transforms.     

Multiplication product formula for associated homogeneous distributions on R

The set of associated homogeneous distributions (AHDs) on R, ??^′(R), consists of the distributional analogues of power‐log functions with domain in R. This set contains the majority of the (one‐dimensional) distributions one typically encounters in physics applications. The recent work done by the author showed that the set ??^′(R) admits a closed convolution structure (??^′(R), *). By combining this structure with the generalized convolution theorem, a distributional multiplication product was defined, resulting in also a closed multiplication structure (??^′(R), .). In this paper, the general multiplication product formula for this structure is derived. Multiplication of AHDs on R is associative, except for critical triple products. These critical products are shown to be non‐associative in a simple and interesting way. The non‐associativity is necessary and sufficient to circumvent Schwartz's impossibility theorem on the multiplication of distributions. Copyright © 2011 John Wiley & Sons, Ltd.     

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

ABSTRACT. The diurnal distribution and abundance dynamics of loafing Glaucous‐winged Gulls (Larus glaucescens) were examined at Protection Island National Wildlife Refuge, Strait of Juan de Fuca, Washington. Asynchronous movement of gulls among three habitat patches dedicated to loafing was modeled as a function of environmental variables using differential equations. Multiple time scale analysis led to the derivation of algebraic models for habitat patch occupancy dynamics. The models were parameterized with hourly census data collected from each habitat patch, and the resulting model predictions were compared with observed census data. A four‐compartment model explained 41% of the variability in the data. Models that predict the dynamics of organism distribution and abundance enhance understanding of the temporal and spatial organization of ecological systems, as well as the decision‐making process in natural resource management.     

NHPP models for categorized software defects

We develop NHPP models to characterize categorized event data, with application to modelling the discovery process for categorized software defects. Conditioning on the total number of defects, multivariate models are proposed for modelling the defects by type. A latent vector autoregressive structure is used to characterize dependencies among the different types. We show how Bayesian inference can be achieved via MCMC procedures, with a posterior prediction‐based L‐measure used for model selection. The results are illustrated for defects of different types found during the System Test phase of a large operating system software development project. Copyright © 2005 John Wiley & Sons, Ltd.     

Lanczos and the Riemannian SVD in information retrieval applications

Variations of the latent semantic indexing (LSI) method in information retrieval (IR) require the computation of singular subspaces associated with the k dominant singular values of a large m × n sparse matrix A, where k?min(m,n). The Riemannian SVD was recently generalized to low‐rank matrices arising in IR and shown to be an effective approach for formulating an enhanced semantic model that captures the latent term‐document structure of the data. However, in terms of storage and computation requirements, its implementation can be much improved for large‐scale applications. We discuss an efficient and reliable algorithm, called SPK‐RSVD‐LSI, as an alternative approach for deriving the enhanced semantic model. The algorithm combines the generalized Riemannian SVD and the Lanczos method with full reorthogonalization and explicit restart strategies. We demonstrate that our approach performs as well as the original low‐rank Riemannian SVD method by comparing their retrieval performance on a well‐known benchmark document collection. Copyright 2004 John Wiley & Sons, Ltd.     

A NEW APPROACH TO SIMULATING FISHERIES DATA FOR POLICY MAKING

The main objective of natural resource management is to create social and economic value while maintaining sustainability. In this paper, we introduce an enhanced method for simulating high‐dimensional time series and apply it to Icelandic fishing resource management data. The methodology can be used in many contexts, but is particularly appropriate for simulating the many complex interactions involved in natural resource management. The simulations can be used to explore the sensitivity of resource management policies to future changes using an affinity parameter. Affinity, qualitatively similar to correlation, is a ordinal measure between –1 and +1 that models one's belief how much the future might behave like, or different from, the past. The main appeal of the method is its reliance on data and relative independence from assumptions about that data. In the paper, we apply it on data on Icelandic cod with encouraging results.     

Forecasting retained earnings of privately held companies with PCA and L1 regression

We use proprietary data collected by SVB Analytics, an affiliate of Silicon Valley Bank, to forecast the retained earnings of privately held companies. Combining methods of principal component analysis (PCA) and L¹/quantile regression, we build multivariate linear models that feature excellent in‐sample fit and strong out‐of‐sample predictive accuracy. The combined PCA and L¹ technique effectively deals with multicollinearity and non‐normality of the data, and also performs favorably when compared against a variety of other models. Additionally, we propose a variable ranking procedure that explains which variables from the current quarter are most predictive of the next quarter's retained earnings. We fit models to the top five variables identified by the ranking procedure and thereby, discover interpretable models with excellent out‐of‐sample performance. Copyright © 2013 John Wiley & Sons, Ltd.     

On an autoregressive model with time-dependent coefficients

Summary As one of the non-stationary time series model, we consider a firstorder autoregressive model in which the autoregressive coefficient is assumed to be a function,f _t (θ), of timet. We establish several assumptions onf _t (θ), not on the terms in the Taylor expansion of log-likelihood function, and show that the estimators of unknown parameters involved inf _t (θ) have strong consistency and asymptotic normality under these assumptions when sample size tends to infinity.     

A PREDICTIVE MODEL TO INFORM ADAPTIVE MANAGEMENT OF DOUBLE‐CRESTED CORMORANTS AND FISHERIES IN MICHIGAN

The proliferation of double‐crested cormorants (DCCOs; Phalacrocorax auritus) in North America has raised concerns over their potential negative impacts on game, cultured and forage fishes, island and terrestrial resources, and other colonial water birds, leading to increased public demands to reduce their abundance. By combining fish surplus production and bird functional feeding response models, we developed a deterministic predictive model representing bird–fish interactions to inform an adaptive management process for the control of DCCOs in multiple colonies in Michigan. Comparisons of model predictions with observations of changes in DCCO numbers under management measures implemented from 2004 to 2012 suggested that our relatively simple model was able to accurately reconstruct past DCCO population dynamics. These comparisons helped discriminate among alternative parameterizations of demographic processes that were poorly known, especially site fidelity. Using sensitivity analysis, we also identified remaining critical uncertainties (mainly in the spatial distributions of fish vs. DCCO feeding areas) that can be used to prioritize future research and monitoring needs. Model forecasts suggested that continuation of existing control efforts would be sufficient to achieve long‐term DCCO control targets in Michigan and that DCCO control may be necessary to achieve management goals for some DCCO‐impacted fisheries in the state. Finally, our model can be extended by accounting for parametric or ecological uncertainty and including more complex assumptions on DCCO–fish interactions as part of the adaptive management process.     

On generating multivariate Poisson data in management science applications

Generating multivariate Poisson random variables is essential in many applications, such as multi echelon supply chain systems, multi‐item/multi‐period pricing models, accident monitoring systems, etc. Current simulation methods suffer from limitations ranging from computational complexity to restrictions on the structure of the correlation matrix, and therefore are rarely used in management science. Instead, multivariate Poisson data are commonly approximated by either univariate Poisson or multivariate Normal data. However, these approximations are often not adequate in practice. In this paper, we propose a conceptually appealing correction for NORTA (NORmal To Anything) for generating multivariate Poisson data with a flexible correlation structure and rates. NORTA is based on simulating data from a multivariate Normal distribution and converting it into an arbitrary continuous distribution with a specific correlation matrix. We show that our method is both highly accurate and computationally efficient. We also show the managerial advantages of generating multivariate Poisson data over univariate Poisson or multivariate Normal data. Copyright © 2011 John Wiley & Sons, Ltd.     

Comparison of strategies when building linear prediction models

In statistical and biometric sciences, one often uses predictive linear models. The initial form of such models is usually obtained by fitting the coefficients of the model to a set of observed data according to the classical least squares method. Newborn models that are obtained in this way will be referred to as raw models. Such raw models are often subject of efforts to improve them as to their predictive performance on external datasets. Several methods can be followed to fine‐tune raw models, thus leading to a variety of model building strategies. In this paper, the idea of so‐called victory rates is introduced to compare the performance of building strategies mutually.Copyright © 2013 John Wiley & Sons, Ltd.     

Mathematical and numerical modellization of large‐scale oceanic waves

This paper is devoted to the theoretical and numerical study of a method which computes the variability of current and density in an oceanic domain. The equations are of Navier–Stokes type for the velocity and of transport‐diffusion type for the density. They are linearized around a given mean circulation and modified by physical assumptions including hydrostatic approximation. The existence and uniqueness of a solution are proved for two sets of equations: first the three‐dimensional problem and then the two‐dimensional cyclic problem derived by assuming a sinusoïdal x‐dependence for the perturbation of the mean flow. The latter corresponds to a modellization of tropical instability waves which are illustrated by the ‘El Nino’ phenomenon. These two problems differ from classical ones because of hydrostatic approximation, boundary conditions imposed by the oceanic domain and complex‐valued functions for the cyclic case. A numerical model is developed for the two‐dimensional cyclic equations. Time discretization is performed by the characteristics method; space discretization uses Q₁ finite elements. Numerical results are presented in a realistic case corresponding to the tropical Pacific Ocean. Copyright © 1999 John Wiley & Sons. Ltd.     

Evaluating flood extent mapping of two hydraulic models, 1D HEC‐RAS and 2D LISFLOOD‐FP in comparison with aerial imagery observations in Gorgan flood plain,Iran

We compared flood mapping techniques using a one‐dimensional (1D) hydraulic model HEC‐RAS and two‐dimensional (2D) LISFLOOD‐FP for a 10‐km reach of Gorgan River in Iran. Both models were run using the same hydrologic input data. The input into the models was a steady discharge of 90 cm, corresponds to a flood peak occurred on March 25, 2012. Flood maps generated using these two models were compared with an observed flood inundation map, using F‐statistic. The roughness coefficients of the models were calibrated by maximizing the value of the F‐statistic. Based on the F‐statistic, LISFLOOD‐FP gives a slightly better result (F = 0.69) than HEC‐RAS (F = 0.67). Visual comparison of the flood extents generated by the two models showed reasonably good agreement. Validation was done using a flood event occurred on May 31, 2014. The LISFLOOD‐FP model gave a better result for validation as well. The 2D model showed more consistency in comparison with the 1D model.     

A new shallow water model with polynomial dependence on depth

In this paper, we study two‐dimensional Euler equations in a domain with small depth. With this aim, we introduce a small non‐dimensional parameter ε related to the depth and we use asymptotic analysis to study what happens when ε becomes small. We obtain a model for ε small that, after coming back to the original domain, gives us a shallow water model that considers the possibility of a non‐constant bottom, and the horizontal velocity has a dependence on z introduced by the vorticity when it is not zero. This represents an interesting novelty with respect to shallow water models found in the literature. We stand out that we do not need to make a priori assumptions about velocity or pressure behaviour to obtain the model. The new model is able to approximate the solutions to Euler equations with dependence on z (reobtaining the same velocities profile), whereas the classic model just obtains the average velocity. Copyright © 2007 John Wiley & Sons, Ltd.