QUASI-POISSON VS. NEGATIVE BINOMIAL REGRESSION: HOW SHOULD WE MODEL OVERDISPERSED COUNT DATA?
Quasi-Poisson and negative binomial regression models have equal numbers of parameters, and either could be used for overdispersed count data. While they often give similar results, there can be striking differences in estimating the effects of covariates. We explain when and why such differences oc...
Main Authors: | , |
---|---|
Format: | Article in Journal/Newspaper |
Language: | unknown |
Published: |
Figshare
2016
|
Subjects: | |
Online Access: | https://dx.doi.org/10.6084/m9.figshare.c.3300023.v1 https://figshare.com/collections/QUASI-POISSON_VS_NEGATIVE_BINOMIAL_REGRESSION_HOW_SHOULD_WE_MODEL_OVERDISPERSED_COUNT_DATA_/3300023/1 |
id |
ftdatacite:10.6084/m9.figshare.c.3300023.v1 |
---|---|
record_format |
openpolar |
spelling |
ftdatacite:10.6084/m9.figshare.c.3300023.v1 2023-05-15T16:33:08+02:00 QUASI-POISSON VS. NEGATIVE BINOMIAL REGRESSION: HOW SHOULD WE MODEL OVERDISPERSED COUNT DATA? Hoef, Jay M. Ver Boveng, Peter L. 2016 https://dx.doi.org/10.6084/m9.figshare.c.3300023.v1 https://figshare.com/collections/QUASI-POISSON_VS_NEGATIVE_BINOMIAL_REGRESSION_HOW_SHOULD_WE_MODEL_OVERDISPERSED_COUNT_DATA_/3300023/1 unknown Figshare https://dx.doi.org/10.1890/07-0043.1 https://dx.doi.org/10.6084/m9.figshare.c.3300023 CC-BY http://creativecommons.org/licenses/by/3.0/us CC-BY Environmental Science Ecology FOS Biological sciences Collection article 2016 ftdatacite https://doi.org/10.6084/m9.figshare.c.3300023.v1 https://doi.org/10.1890/07-0043.1 https://doi.org/10.6084/m9.figshare.c.3300023 2021-11-05T12:55:41Z Quasi-Poisson and negative binomial regression models have equal numbers of parameters, and either could be used for overdispersed count data. While they often give similar results, there can be striking differences in estimating the effects of covariates. We explain when and why such differences occur. The variance of a quasi-Poisson model is a linear function of the mean while the variance of a negative binomial model is a quadratic function of the mean. These variance relationships affect the weights in the iteratively weighted least-squares algorithm of fitting models to data. Because the variance is a function of the mean, large and small counts get weighted differently in quasi-Poisson and negative binomial regression. We provide an example using harbor seal counts from aerial surveys. These counts are affected by date, time of day, and time relative to low tide. We present results on a data set that showed a dramatic difference on estimating abundance of harbor seals when using quasi-Poisson vs. negative binomial regression. This difference is described and explained in light of the different weighting used in each regression method. A general understanding of weighting can help ecologists choose between these two methods. Article in Journal/Newspaper harbor seal DataCite Metadata Store (German National Library of Science and Technology) |
institution |
Open Polar |
collection |
DataCite Metadata Store (German National Library of Science and Technology) |
op_collection_id |
ftdatacite |
language |
unknown |
topic |
Environmental Science Ecology FOS Biological sciences |
spellingShingle |
Environmental Science Ecology FOS Biological sciences Hoef, Jay M. Ver Boveng, Peter L. QUASI-POISSON VS. NEGATIVE BINOMIAL REGRESSION: HOW SHOULD WE MODEL OVERDISPERSED COUNT DATA? |
topic_facet |
Environmental Science Ecology FOS Biological sciences |
description |
Quasi-Poisson and negative binomial regression models have equal numbers of parameters, and either could be used for overdispersed count data. While they often give similar results, there can be striking differences in estimating the effects of covariates. We explain when and why such differences occur. The variance of a quasi-Poisson model is a linear function of the mean while the variance of a negative binomial model is a quadratic function of the mean. These variance relationships affect the weights in the iteratively weighted least-squares algorithm of fitting models to data. Because the variance is a function of the mean, large and small counts get weighted differently in quasi-Poisson and negative binomial regression. We provide an example using harbor seal counts from aerial surveys. These counts are affected by date, time of day, and time relative to low tide. We present results on a data set that showed a dramatic difference on estimating abundance of harbor seals when using quasi-Poisson vs. negative binomial regression. This difference is described and explained in light of the different weighting used in each regression method. A general understanding of weighting can help ecologists choose between these two methods. |
format |
Article in Journal/Newspaper |
author |
Hoef, Jay M. Ver Boveng, Peter L. |
author_facet |
Hoef, Jay M. Ver Boveng, Peter L. |
author_sort |
Hoef, Jay M. Ver |
title |
QUASI-POISSON VS. NEGATIVE BINOMIAL REGRESSION: HOW SHOULD WE MODEL OVERDISPERSED COUNT DATA? |
title_short |
QUASI-POISSON VS. NEGATIVE BINOMIAL REGRESSION: HOW SHOULD WE MODEL OVERDISPERSED COUNT DATA? |
title_full |
QUASI-POISSON VS. NEGATIVE BINOMIAL REGRESSION: HOW SHOULD WE MODEL OVERDISPERSED COUNT DATA? |
title_fullStr |
QUASI-POISSON VS. NEGATIVE BINOMIAL REGRESSION: HOW SHOULD WE MODEL OVERDISPERSED COUNT DATA? |
title_full_unstemmed |
QUASI-POISSON VS. NEGATIVE BINOMIAL REGRESSION: HOW SHOULD WE MODEL OVERDISPERSED COUNT DATA? |
title_sort |
quasi-poisson vs. negative binomial regression: how should we model overdispersed count data? |
publisher |
Figshare |
publishDate |
2016 |
url |
https://dx.doi.org/10.6084/m9.figshare.c.3300023.v1 https://figshare.com/collections/QUASI-POISSON_VS_NEGATIVE_BINOMIAL_REGRESSION_HOW_SHOULD_WE_MODEL_OVERDISPERSED_COUNT_DATA_/3300023/1 |
genre |
harbor seal |
genre_facet |
harbor seal |
op_relation |
https://dx.doi.org/10.1890/07-0043.1 https://dx.doi.org/10.6084/m9.figshare.c.3300023 |
op_rights |
CC-BY http://creativecommons.org/licenses/by/3.0/us |
op_rightsnorm |
CC-BY |
op_doi |
https://doi.org/10.6084/m9.figshare.c.3300023.v1 https://doi.org/10.1890/07-0043.1 https://doi.org/10.6084/m9.figshare.c.3300023 |
_version_ |
1766022852894523392 |