Anders Paarup Nielsen

AD Model Builder: using automatic differentiation for statistical inference of highly parameterized complex nonlinear models

Many criteria for statistical parameter estimation, such as maximum likelihood, are formulated as a nonlinear optimization problem. Automatic Differentiation Model Builder (ADMB) is a programming framework based on automatic differentiation, aimed at highly nonlinear models with a large number of parameters. The benefits of using AD are computational efficiency and high numerical accuracy, both crucial in many practical problems. We describe the basic components and the underlying philosophy of ADMB, with an emphasis on functionality found in no other statistical software. One example of such a feature is the generic implementation of Laplace approximation of high-dimensional integrals for use in latent variable models. We also review the literature in which ADMB has been used, and discuss future development of ADMB as an open source project. Overall, the main advantages of ADMB are flexibility, speed, precision, stability and built-in methods to quantify uncertainty.

Understanding dynamic capabilities through knowledge management

Purpose – This paper sets out to integrate research on knowledge management with the dynamic capabilities approach. This paper will add to the understanding of dynamic capabilities by demonstrating that dynamic capabilities can be seen as composed of concrete and well‐known knowledge management activities.Design/methodology/approach – This paper is based on a literature review focusing on key knowledge management processes and activities as well as the concept of dynamic capabilities, the paper connects these two approaches. The analysis is centered on knowledge management activities which then are compiled into dynamic capabilities.Findings – In the paper eight knowledge management activities are identified; knowledge creation, acquisition, capture, assembly, sharing, integration, leverage, and exploitation. These activities are assembled into the three dynamic capabilities of knowledge development, knowledge (re)combination, and knowledge use. The dynamic capabilities and the associated knowledge manage...

Preventing the collapse of the Baltic cod stock through an ecosystem-based management approach

Worldwide a number of fish stocks have collapsed because of overfishing and climate-induced ecosystem changes. Developing ecosystem-based fisheries management (EBFM) to prevent these catastrophic events in the future requires ecological models incorporating both internal food-web dynamics and external drivers such as fishing and climate. Using a stochastic food-web model for a large marine ecosystem (i.e., the Baltic Sea) hosting a commercially important cod stock, we were able to reconstruct the history of the stock. Moreover we demonstrate that in hindsight the collapse could only have been avoidable by adapting fishing pressure to environmental conditions and food-web interactions. The modeling approach presented here represents a significant advance for EBFM, the application of which is important for sustainable resource management in the future.

Ecological forecasting under climate change: the case of Baltic cod

Good decision making for fisheries and marine ecosystems requires a capacity to anticipate the consequences of management under different scenarios of climate change. The necessary ecological forecasting calls for ecosystem-based models capable of integrating multiple drivers across trophic levels and properly including uncertainty. The methodology presented here assesses the combined impacts of climate and fishing on marine food-web dynamics and provides estimates of the confidence envelope of the forecasts. It is applied to cod (Gadus morhua) in the Baltic Sea, which is vulnerable to climate-related decline in salinity owing to both direct and indirect effects (i.e. through species interactions) on early-life survival. A stochastic food web-model driven by regional climate scenarios is used to produce quantitative forecasts of cod dynamics in the twenty-first century. The forecasts show how exploitation would have to be adjusted in order to achieve sustainable management under different climate scenarios.

Strategies for fitting nonlinear ecological models in R, AD Model Builder, and BUGS

1. Ecologists often use nonlinear fitting techniques to estimate the parameters of complex ecological models, with attendant frustration. This paper compares three open-source model fitting tools and discusses general strategies for defining and fitting models. 2. R is convenient and (relatively) easy to learn, AD Model Builder is fast and robust but comes with a steep learning curve, while BUGS provides the greatest flexibility at the price of speed. 3. Our model-fitting suggestions range from general cultural advice (where possible, use the tools and models that are most common in your subfield) to specific suggestions about how to change the mathematical description of models to make them more amenable to parameter estimation. 4. A companion web site (https://groups.nceas.ucsb.edu/nonlinear-modeling/projects) presents detailed examples of application of the three tools to a variety of typical ecological estimation problems; each example links both to a detailed project report and to full source code and data.

Living carbocationic polymerization of isobutylene and synthesis of ABA block copolymers by conventional laboratory techniques

SummaryLiving polymerization of isobutylene (IB) and subsequent controlled synthesis of ABA block copolymers, such as poly(styrene-b-isobutylene-b-styrene) (PSt-PIB-PSt) and poly(p-methylstyrene-b-isobutylene-b-p-methylstyrene) (PpMeSt-PIB-PpMeSt), have been carried out by a simple and inexpensive conventional laboratory technique. The homo- and block copolymers obtained by using this technique have exhibited excellent molecular weight control and low polydispersity indexes. The living nature of IB polymerization has been demonstrated by the incremental monomer addition (IMA) method with the dicumyl methyl ether (DiCumOMe)/TiCl4 initiating system in the presence of 2,5-di-tert-butylpyridine (DtBP) proton trap. PSt-PIB-PSt and PpMeSt-PIB-PpMeSt block copolymers have been synthesized by sequential monomer addition: first living difunctional polyisobutylene (PIB) midsegment was prepared by difunctional initiator, then the second monomer was added to the charge. High blocking efficiencies and desired block copolymer structures have been obtained.

Implementing lean in surgery – lessons and implications

The principles of lean production originating from the Toyota production system has spread from manufacturing to healthcare. Needless to say, this raises concern whether such principles are actually applicable where the product are humans in need of medical care and what are the consequences for the medical staff. The literature on lean does not suggest that lean should not be applicable in healthcare. Still this paper begins by discussing if the principles of lean from a theoretical perspective can be expected to work in healthcare. It is found that that it will be useful in certain types of settings but it is not a universal tool that may be applied to all aspects of healthcare. A case of lean in a surgical ward is presented and it is observed that patients have been split in to two flows, which can be described as a lean and a normal flow.

Modeling Zero-Inflated Count Data With glmmTMB

Ecological phenomena are often measured in the form of count data. These data can be analyzed using generalized linear mixed models (GLMMs) when observations are correlated in ways that require random effects. However, count data are often zero-inflated, containing more zeros than would be expected from the standard error distributions used in GLMMs, e.g., parasite counts may be exactly zero for hosts with effective immune defenses but vary according to a negative binomial distribution for non-resistant hosts. We present a new R package, glmmTMB, that increases the range of models that can easily be fitted to count data using maximum likelihood estimation. The interface was developed to be familiar to users of the lme4 R package, a common tool for fitting GLMMs. To maximize speed and flexibility, estimation is done using Template Model Builder (TMB), utilizing automatic differentiation to estimate model gradients and the Laplace approximation for handling random effects. We demonstrate glmmTMB and compare it to other available methods using two ecological case studies. In general, glmmTMB is more flexible than other packages available for estimating zero-inflated models via maximum likelihood estimation and is faster than packages that use Markov chain Monte Carlo sampling for estimation; it is also more flexible for zero-inflated modelling than INLA, but speed comparisons vary with model and data structure. Our package can be used to fit GLMs and GLMMs with or without zero-inflation as well as hurdle models. By allowing ecologists to quickly estimate a wide variety of models using a single package, glmmTMB makes it easier to find appropriate models and test hypotheses to describe ecological processes.

Fishing power increases from technological development in the Faroe Islands longline fishery

During an effort-regulated period from 1996 to 2002, unregistered annual increases of 0.3% of hooks fished per day were demonstrated for the Faroe Islands longline fishery. However, annual increases were higher (1.5%) during a preceding total allowable catch regulated period, thereby invalidating an expectation of the 1996 shift in regulations (output to input control) to have induced increases in the number of hooks set per day. Underlying this result is a substantial increase in total yearly effort (fishing days) and a shift in targeting behaviour from secondary to primary (high value) target species in response to the transition from output to input control. Interview data on technology were combined with logbook data and analysed with generalized linear modelling to demonstrate haddock (Melanogrammus aeglefinus) and Atlantic cod (Gadus morhua) catch-per-unit-effort increases of 51% and 26%, respectively, following the introduction of skewed hooks and swivel line. The technological introductions were n...

Lessons from a Prototype Geolocation Problem

This paper establishes simple and general expressions for the accuracy of geolocation, which may be obtained by optimal filtering of measurements from archival tags. We investigate an idealized geolocation problem where the animal performs a random walk. We derive simple closed-form expressions for the steady-state variance and for the characteristic time scale of the filter, i.e. the smoothing horizon. This leads to temporal and spatial scales defining the limit of resolution and explains the difference between what can be obtained for fast-moving and slow-moving animals. Using frequency-domain methods, we consider the effect of adding additional sensors, and examine the substitution of the random walk model with anomalous diffusion, e.g. a Levy flight. We also discuss time variations in the accuracy near start and end of the time series, and due to holes in the data stream which e.g. arise in the tidal method for geolocation when the animal is pelagic. Our results are particularly useful to the planning of a tagging study, because our estimates of accuracy can be computed using only three parameters: the swimming speed of the animal, the sample interval, and the variance on the measurement error.