Generalized additive modelling and zero inflated count data

Barry, Simon C.; Welsh, A. H.

doi:10.1016/s0304-3800(02)00194-1

Cited by 258 publications

(178 citation statements)

References 5 publications

Supporting

Mentioning

177

Contrasting

Order By: Relevance

“…Lambert (1992) provided techniques for modeling data with a zero-inflated Poisson (ZIP) model. Incorporating zero-inflated likelihoods in spatially explicit models is described in Agarwal et al (2002), Barry and Welsh (2002), Fahrmeir and Echavarria (in press), Gschlobl and Czado (2006), Rathbun and Fei (2006), Rigby and Stasinopoulos (2005) and in general (not spatial) in Hall (2000), Lambert (1992), and Li et al (1999 SemiPar via a space-filling algorithm (Ruppert et al 2003) with the default number of 50 knots.…”

Section: Methodsmentioning

confidence: 99%

Proceedings of the eighth annual forest inventory and analysis symposium

McRoberts¹,

Reams²,

Deusen³

et al. 2009

View full text Add to dashboard Cite

Section: Methodsmentioning

confidence: 99%

Proceedings of the eighth annual forest inventory and analysis symposium

McRoberts¹,

Reams²,

Deusen³

et al. 2009

View full text Add to dashboard Cite

“…The reason for this is that the data are zero inflated. The negative binomial or the Poisson model assumption is not an adequate approximation to the conditional distribution of the data (Barry and Welsh, 2002). One possible solution to this could be to model the zero-observations separately as done in (Barry and Welsh, 2002;Borchers et al, 1997).…”

Section: Photographic Surveys -Abundance and Uncertainty Estimation Bmentioning

confidence: 99%

“…Due to this we experienced that the GAM method was numerically unstable. Dealing with zero inflated data is a known issue for the GAM method and has been reported elsewhere (Barry and Welsh, 2002). The main problem is that the underlying assumption that the counts in the photos are negative binomial distributed is not met.…”

Section: Conventional Strip Transect Analysis Versus Gam Analysismentioning

confidence: 99%

Estimation of Pup Production of Hooded and Harp Seals in the Greenland Sea in 2007: Reducing Uncertainty Using Generalized Additive Models

Øigærd¹,

Haug²,

Nilssen³

et al. 2010

J. Northw. Atl. Fish. Sci.

View full text Add to dashboard Cite

Øigård, T. A., T. Haug, K. T. Nilssen, and A. B. Salberg. 2010. Estimation of pup production of hooded and harp seals in the Greenland Sea in 2007: reducing uncertainty using generalized additive models. J. Northw. Atl. Fish. Sci., 42: 103-123. doi:10.2960/J.v42.m642 AbstractThe pup production of the Greenland Sea populations of hooded and harp seals were assessed in aerial surveys using two aircrafts for reconnaissance flights and photographic surveys along transects over the whelping areas from 14 March to 3 April 2007. One helicopter, operated from the applied expedition vessel, flew reconnaissance flights, monitored the distribution of seal patches and performed age-staging of the pups. The total estimate of hooded seal pup production was 16 140 (SE = 2 140, CV = 13.3%), which is similar to an estimate obtained for comparable surveys in 2005. The total pup production estimate obtained for harp seals was 110 530 (SE = 27 680, CV = 25.0%), which is slightly higher than an estimate obtained for a similar survey in 2002. The pup production and the uncertainty of the pup production estimate were estimated using a standard method for analyzing this type of survey data and a recently developed method that utilized Generalized Additive Models (GAMs). Using the two estimation methods on data from all three surveys (2002, 2005, 2007), comparable estimates of pup production were obtained. In scenarios where pups were clustered, the estimated uncertainty of the pup production estimate was much lower for the GAM method than for the conventional method. This resulted in a considerable reduction of the estimated coefficient of variation. In scenarios where pups were uniformly distributed, both methods performed the same.

show abstract

“…We therefore followed the approach recommended by Barry and Welsh (2002) chosen as it proved to be the depth of maximal krill biomass density, and as such gave an indication of currents experienced by the greater part of the krill. Furthermore, the patterns described below concerning the association of krill biomass with currents and shear at 150 m were highly similar to the unreported associations of biomass with currents and shear at shallower depths where currents were stronger.…”

Section: 25b Statistical Analysismentioning

confidence: 99%

Distribution, patchiness, and behavior of Antarctic zooplankton, assessed using multi-frequency acoustic techniques

Lawson¹

2006

View full text Add to dashboard Cite

The physical and biological forces that drive zooplankton distribution and patchiness in an antarctic continental shelf region were examined, with particular emphasis on the Antarctic krill, Euphausia superba. This was accomplished by the application of acoustic, video, and environmental sensors during surveys of the region in and around Marguerite Bay, west of the Antarctic Peninsula, in the falls and winters of 2001 and 2002. An important component of the research involved the development and verification of methods for extracting estimates of ecologically-meaningful quantities from measurements of scattered sound. The distribution of acoustic volume backscattering at the single frequency of 120 kHz was first examined as an index of the overall biomass of zooplankton. Distinct spatial and seasonal patterns were observed that coincided with advective features. Improved parameterization was then achieved for a theoretical model of Antarctic krill target strength, the quantity necessary in scaling measurements of scattered sound to estimates of abundance, through direct measurement of all necessary model parameters for krill sampled in the study region and survey period. Methods were developed for identifying and delineating krill aggregations, allowing the distribution of krill to be distinguished from that of the overall zooplankton community. Additional methods were developed and verified for estimating the length, abundance, and biomass of krill in each acoustically-identified aggregation. These methods were applied to multifrequency acoustic survey data, demonstrating strong seasonal, inter-annual, and spatial variability in the distribution of krill biomass. Highest biomass was consistently associated with regions close to land where temperatures at depth were cool. Finally, the morphology, internal structure, and vertical position of individual krill aggregations were examined. The observed patterns of variability in aggregation characteristics between day and night, regions of high versus low food availability, and in the presence or absence of predators, together reinforced the conclusion that aggregation and diel vertical migration represent strategies to avoid visual predators, while also allowing the krill access to shallowly-distributed food resources. The various findings of this work have important implications to the fields of zooplankton acoustics and Antarctic krill ecology, especially in relation to the interactions of the krill with its predators.

show abstract

Generalized additive modelling and zero inflated count data

Cited by 258 publications

References 5 publications

Proceedings of the eighth annual forest inventory and analysis symposium

Proceedings of the eighth annual forest inventory and analysis symposium

Estimation of Pup Production of Hooded and Harp Seals in the Greenland Sea in 2007: Reducing Uncertainty Using Generalized Additive Models

Distribution, patchiness, and behavior of Antarctic zooplankton, assessed using multi-frequency acoustic techniques

Contact Info

Product

Resources

About