A Simulation Study to Evaluate Survey Designs and Assessment Models for Estimation of Dungeness Crab (Cancer magister) Softshell Periods

Softshell Dungeness Crabs have inferior meat quality and are vulnerable to handling by harvesters; therefore, knowing when softshell periods occur is important for managing Dungeness Crab fisheries. A computer simulation was used to study the effectiveness of several survey designs and statistical models for estimating softshell periods which normally would be construed from crab shell condition data obtained from trap surveys. Survey designs varied in the number of years of data collection (1, 3, 5 or 10 years) and by the number and arrangement of sampling events per year. Three statistical models, including standardized catch-per-unit-effort (SCPUE), hierarchical, and generalized additive, were tested using catch-per-unit-effort data (CPUEs) or CPUE- transformed data. CPUEs were standardised by dividing CPUE estimates by the maximum CPUE obtained in the sample year, and then transformed using the complementary log-log function. In the hierarchical model, CPUEs were modelled using a lognormal distribution, assuming the expected logarithms of CPUEs are a quadratic function of days plus a random normal error. CPUE-transformed data were modelled using a normal distribution, assuming expected values are a quadratic function of days in the SCPUE model or a spline smooth function of days in the generalized additive model. Results suggest the best survey design requires a relatively high number (6 or 11) of sampling events during several key consecutive months which contain the softshell period, and fewer sampling events during those months when softshell crab abundance is low. A minimum 3 years of data collection is required to produce reliable outputs. The hierarchical model performs best, slightly better than the SCPUE model. Use of the generalized additive model is not recommended.

Download Full-text

Hierarchical generalized additive models: an introduction with mgcv

10.7287/peerj.preprints.27320v1 ◽

2018 ◽

Cited By ~ 1

Author(s):

Eric J Pedersen ◽

David L. Miller ◽

Gavin L. Simpson ◽

Noam Ross

Keyword(s):

Hierarchical Model ◽

Generalized Additive Model ◽

Generalized Additive Models ◽

Additive Model ◽

Additive Models ◽

Complex Structures ◽

Ecological Data ◽

Nonlinear Functional ◽

Functional Relationships ◽

Statistical Issues

In this paper, we discuss an extension to two popular approaches to modelling complex structures in ecological data: the generalized additive model (GAM) and the hierarchical model (HGLM). The hierarchical GAM (HGAM), allows modelling of nonlinear functional relationships between covariates and outcomes where the shape of the function itself varies between different grouping levels. We describe the theoretical connection between these models, HGLMs and GAMs, explain how to model different assumptions about the degree of inter-group variability in functional response, and show how HGAMs can be readily fitted using existing GAM software, the mgcv package in R. We also discuss computational and statistical issues with fitting these models, and demonstrate how to fit HGAMs on example data.

Download Full-text

Modelling Cyclists’ Multi-Exposure to Air and Noise Pollution with Low-Cost Sensors—The Case of Paris

Atmosphere ◽

10.3390/atmos11040422 ◽

2020 ◽

Vol 11 (4) ◽

pp. 422

Author(s):

Jérémy Gelb ◽

Philippe Apparicio

Keyword(s):

Statistical Models ◽

Noise Exposure ◽

Generalized Additive Model ◽

Low Cost ◽

Additive Model ◽

Noise Pollution ◽

Ventilation Rate ◽

Mixed Effects ◽

Mobile Data ◽

Mobile Data Collection

Cyclists are particularly exposed to air and noise pollution because of their higher ventilation rate and their proximity to traffic. However, few studies have investigated their multi-exposure and have taken into account its real complexity in building statistical models (nonlinearity, pseudo replication, autocorrelation, etc.). We propose here to model cyclists’ exposure to air and noise pollution simultaneously in Paris (France). Specifically, the purpose of this study is to develop a methodology based on an extensive mobile data collection using low-cost sensors to determine which factors of the urban micro-scale environment contribute to cyclists’ multi-exposure and to what extent. To this end, we developed a conceptual framework to define cyclists’ multi-exposure and applied it to a multivariate generalized additive model with mixed effects and temporal autocorrelation. The results show that it is possible to reduce cyclists’ multi-exposure by adapting the planning and development practices of cycling infrastructure, and that this reduction can be substantial for noise exposure.

Download Full-text

Relationship between catch, effort, CPUE and local abundance for non-target species, such as billfishes, caught by Indian Ocean longline fisheries

Marine and Freshwater Research ◽

10.1071/mf01268 ◽

2003 ◽

Vol 54 (4) ◽

pp. 383 ◽

Cited By ~ 13

Author(s):

Alain Fonteneau ◽

Nicolas Richard

Keyword(s):

Indian Ocean ◽

Generalized Additive Model ◽

Additive Model ◽

Fishing Effort ◽

Potential Bias ◽

Target Species ◽

Catch Per Unit Effort ◽

The Indian Ocean ◽

Longline Fisheries ◽

The Relationship

This paper analyses the local relationship between effort, catches, catch per unit effort (CPUE) and abundance of target species (such as tunas) and of non-target species (such as billfishes). The Indian Ocean longline fisheries are taken as an example. This paper evaluates the potential bias in the relationship between local CPUE and abundance when fisheries are increasing their fishing effort. A presentation of the Indian Ocean longline fisheries is carried out. A statistical analysis of CPUE is conducted using a generalized additive model which tends to indicate that the local effort is an important component in the statistical behaviour of the local CPUE. A migratory model in which both resources and fisheries are mobile was built. This model simulates the combined exploitation of two species, a target and a bycatch species, both fished at increasing intensity. This model confirms the potential bias as a result of the concentration of fishing effort in areas of high density of the target species. It also suggests that the CPUE of bycatch species may be more heavily biased because of their status. It is recommended that local fishing efforts should preferably be taken into account in order to calculate the CPUE of both target and non-target species.

Download Full-text

Ambient Sulfate Trends and the Influence of Meteorology

Journal of Applied Meteorology ◽

10.1175/jam2307.1 ◽

2005 ◽

Vol 44 (11) ◽

pp. 1745-1760 ◽

Cited By ~ 5

Author(s):

Stephen F. Mueller

Keyword(s):

Statistical Models ◽

Generalized Additive Model ◽

Additive Model ◽

Total Sulfur ◽

Eastern United States ◽

Sulfur Species ◽

Sulfate Sulfur ◽

Average Change ◽

Meteorological Influences ◽

Local Deposition

Abstract Data on atmospheric levels of sulfur dioxide (SO2) and sulfate were examined to quantify changes since 1989. Changes in sulfur species were adjusted to account for meteorological variability. Adjustments were made using meteorological variables expressed in terms of their principal components that were used as predictors in statistical models. Several models were tested. A generalized additive model (GAM)—based in part on nonparametric, locally smoothed predictor functions—computed the greatest association between sulfate and the meteorological predictors. Sulfate trends estimated after a GAM-based adjustment for weather-related influences were found to be primarily downward across the eastern United States by as much as 6.7% per year (average of −2.6% per year), but large spatial variability was noted. The most conspicuous characteristic in the trends was over portions of the Appalachian Mountains where very small (average = −1.6% per year) and often insignificant sulfate changes were found. The Appalachian region also experienced a tendency, after removing meteorological influences, for increases in the ratio RS of sulfate sulfur to total sulfur. Before 1991, this ratio averaged 0.33 across all sites. Appalachian increases in RS were equivalent to 0.07 during 1989–2001 (significant for most sites at the 0.05 level), or nearly 2 times the average change at the other sites. This suggests that conditions over the Appalachians became notably more efficient at oxidizing SO2 into sulfate. Alternatively, subtle changes in local deposition patterns occurred, preferentially in and near mountainous monitoring sites, that changed the SO2–sulfate balance.

Download Full-text

Hierarchical generalized additive models: an introduction with mgcv

10.7287/peerj.preprints.27320 ◽

2018 ◽

Cited By ~ 1

Author(s):

Eric J Pedersen ◽

David L. Miller ◽

Gavin L. Simpson ◽

Noam Ross

Keyword(s):

Hierarchical Model ◽

Generalized Additive Model ◽

Generalized Additive Models ◽

Additive Model ◽

Additive Models ◽

Complex Structures ◽

Ecological Data ◽

Nonlinear Functional ◽

Functional Relationships ◽

Statistical Issues

Download Full-text

GAMLSS for high-variability data: an application to liver fibrosis case

The International Journal of Biostatistics ◽

10.1515/ijb-2019-0113 ◽

2020 ◽

Vol 0 (0) ◽

Author(s):

Andrea Marletta ◽

Mariangela Sciandra

Keyword(s):

Risk Factors ◽

Liver Fibrosis ◽

Statistical Models ◽

Generalized Additive Model ◽

Additive Model ◽

Finite Mixture Model ◽

High Variability ◽

Finite Mixture ◽

Potential Risk Factors ◽

Variability Data

AbstractThis article aims to provide rigorous and convenient statistical models for dealing with high-variability phenomena. The presence of discrepance in variance represents a substantial issue when it is not possible to reduce variability before analysing the data, leading to the possibility to estimate an inadequate model. In this paper, the application of Generalized Additive Model for Location, Scale and Shape (GAMLSS) and the use of finite mixture model for GAMLSS will be proposed as a solution to the problem of overdispersion. An application to Liver fibrosis data is illustrated in order to identify potential risk factors for patients, which could determine the presence of the disease but also its levels of severity.

Download Full-text

CLIMATE-DRIVEN STATISTICAL MODELS AS EFECTIVE PREDICTIONS OF LOCAL DENGUE INDICENCE IN COSTA RICA: A GENERALIZED ADDITIVE MODEL AND RANDOM FOREST APPROACH

Revista de Matemática Teoría y Aplicaciones ◽

10.15517/rmta.v27i1.39931 ◽

2019 ◽

Vol 27 (1) ◽

pp. 1-21

Author(s):

PAOLA VÁSQUEZ ◽

ANTONIO LORÍA ◽

FABIO SÁNCHEZ ◽

LUIS ALBERTO BARBOZA

Keyword(s):

Public Health ◽

Relative Risk ◽

Costa Rica ◽

Random Forest ◽

Prediction Model ◽

Statistical Models ◽

Generalized Additive Model ◽

Additive Model ◽

Climate Data ◽

Tropical Country

Climate has been an important factor in shaping the distribution and incidence of dengue cases in tropical and subtropical countries. In Costa Rica, a tropical country with distinctive micro-climates, dengue has been endemic since its introduction in 1993, inflicting substantial economic, social, and public health repercussions. Using the number of dengue reported cases and climate data from 2007-2017, we fitted a prediction model applying a Generalized Additive Model (GAM) and Random Forest (RF) approach, which allowed us to retrospectively predict the relative risk of dengue in five climatological diverse municipalities around the country.

Download Full-text