scholarly journals 7. Fixed-Effects Negative Binomial Regression Models

2002 ◽  
Vol 32 (1) ◽  
pp. 247-265 ◽  
Author(s):  
Paul D. Allison ◽  
Richard P. Waterman
Keyword(s):  
Fixed Effects ◽  
Negative Binomial ◽  
Negative Binomial Regression ◽  
Alternative Methods ◽  
Incidental Parameters ◽  
Binomial Regression ◽  
Downward Bias ◽  
Over Dispersion ◽  
Poisson Estimator ◽  
Deviance Statistic

This paper demonstrates that the conditional negative binomial model for panel data, proposed by Hausman, Hall, and Griliches (1984), is not a true fixed-effects method. This method—which has been implemented in both Stata and LIMDEP—does not in fact control for all stable covariates. Three alternative methods are explored. A negative multinomial model yields the same estimator as the conditional Poisson estimator and hence does not provide any additional leverage for dealing with over-dispersion. On the other hand, a simulation study yields good results from applying an unconditional negative binomial regression estimator with dummy variables to represent the fixed effects. There is no evidence for any incidental parameters bias in the coefficients, and downward bias in the standard error estimates can be easily and effectively corrected using the deviance statistic. Finally, an approximate conditional method is found to perform at about the same level as the unconditional estimator.

Managing Inflation

2016 ◽  
Vol 63 (1) ◽  
pp. 77-87 ◽  
Author(s):  
William H. Fisher ◽  
Stephanie W. Hartwell ◽  
Xiaogang Deng
Keyword(s):  
Count Data ◽  
Regression Models ◽  
Negative Binomial ◽  
Negative Binomial Regression ◽  
Excess Zeros ◽  
Binomial Regression ◽  
Negative Binomial Models ◽  
Using Data ◽  
Over Dispersion ◽  
Binomial Models

Poisson and negative binomial regression procedures have proliferated, and now are available in virtually all statistical packages. Along with the regression procedures themselves are procedures for addressing issues related to the over-dispersion and excessive zeros commonly observed in count data. These approaches, zero-inflated Poisson and zero-inflated negative binomial models, use logit or probit models for the “excess” zeros and count regression models for the counted data. Although these models are often appropriate on statistical grounds, their interpretation may prove substantively difficult. This article explores this dilemma, using data from a study of individuals released from facilities maintained by the Massachusetts Department of Correction.

scholarly journals NATO between Exclusivity and Inclusivity: Measuring NATO’s Partnerships

2021 ◽  
pp. 1-32
Author(s):  
Branislav Mičko
Keyword(s):  
Human Rights ◽  
Fixed Effects ◽  
International Institutions ◽  
Negative Binomial ◽  
Negative Binomial Regression ◽  
Original Dataset ◽  
Strategic Approach ◽  
Binomial Regression ◽  
The Given ◽  
The Relationship

Building on an original dataset, this article focuses on the interactions between NATO and its declared worldwide partners. It argues that the analysis of these interactions can reveal NATO’s strategic approach to partnerships, but it can also provide a tool for its classification as an organisation that is either exclusive – defined by the focus on defence of its members, or inclusive – emphasising the global protection of democracies and human rights. The relationship between types of interactions and NATO categorisation is estimated using an unconditional negative binomial regression with fixed effects as well as a within-between (hybrid) model. Furthermore, they are illustrated on two brief case studies of Sweden and Japan. The results of the study suggest that NATO engages primarily with countries that are powerful relative to their neighbourhood, even though they are not the most powerful among the partners. The given country’s level of democracy, integration into the international institutions, and stability, do not seem to play any overarching role here.

scholarly journals Retracted: Analysis of the Strength of Legal Firearms Restrictions for Perpetrators of Domestic Violence and Their Associations With Intimate Partner Homicide

2017 ◽  
Vol 187 (7) ◽  
pp. 1449-1455 ◽  
Author(s):  
April M Zeoli ◽  
Alexander McCourt ◽  
Shani Buggs ◽  
Shannon Frattaroli ◽  
David Lilley ◽  
...  
Keyword(s):  
Domestic Violence ◽  
Fixed Effects ◽  
Negative Binomial ◽  
State Level ◽  
Negative Binomial Regression ◽  
Intimate Partner ◽  
Restraining Order ◽  
Level Data ◽  
Binomial Regression ◽  
Firearms Restrictions

Abstract In this research, we estimate the association of firearm restrictions for domestic violence offenders with intimate partner homicides (IPHs) on the basis of the strength of the policies. We posit that the association of firearm laws with IPHs depends on the following characteristics of the laws: 1) breadth of coverage of high-risk individuals and situations restricted; 2) power to compel firearm surrender or removal from persons prohibited from having firearms; and 3) systems of accountability that prevent those prohibited from doing so from obtaining guns. We conducted a quantitative policy evaluation using annual state-level data from 1980 through 2013 for 45 US states. Based on the results of a series of robust, negative binomial regression models with state fixed effects, domestic violence restraining order firearm-prohibition laws are associated with 10% reductions in IPH. Statistically significant protective associations were evident only when restraining order prohibitions covered dating partners (−11%) and ex parte orders (−12%). Laws prohibiting access to those convicted of nonspecific violent misdemeanors were associated with a 24% reduction in IPH rates; there was no association when prohibitions were limited to domestic violence. Permit-to-purchase laws were associated with 10% reductions in IPHs. These findings should inform policymakers considering laws to maximize protections against IPH.

scholarly journals WITER: A powerful method for the estimation of cancer-driver genes using a weighted iterative regression accurately modelling background mutation rate

2018 ◽  
Author(s):  
Lin Jiang ◽  
Jingjing Zheng ◽  
Johnny Sheung Him Kwan ◽  
Sheng Dai ◽  
Cong Li ◽  
...  
Keyword(s):  
Negative Binomial ◽  
Negative Binomial Regression ◽  
Alternative Methods ◽  
Driver Mutations ◽  
Driver Genes ◽  
Cancer Driver ◽  
Binomial Regression ◽  
Background Mutation Rate ◽  
Technical Advances ◽  
Cancer Driver Genes

AbstractGenomic identification of driver mutations and genes in cancer cells are critical for precision medicine. Due to difficulty in modeling distribution of background mutations, existing statistical methods are often underpowered to discriminate driver genes from passenger genes. Here we propose a novel statistical approach, weighted iterative zero-truncated negative-binomial regression (WITER), to detect cancer-driver genes showing an excess of somatic mutations. By solving the problem of inaccurately modeling background mutations, this approach works even in small or moderate samples. Compared to alternative methods, it detected more significant and cancer-consensus genes in all tested cancers. Applying this approach, we estimated 178 driver genes in 26 different cancers types. In silico validation confirmed 90.5% of predicted genes as likely known drivers and 7 genes unique for individual cancers as likely new drivers. The technical advances of WITER enable the detection of driver genes in TCGA datasets as small as 30 subjects, rescuing more genes missed by alternative tools.

scholarly journals WITER: a powerful method for estimation of cancer-driver genes using a weighted iterative regression modelling background mutation counts

2019 ◽  
Vol 47 (16) ◽  
pp. e96-e96 ◽  
Author(s):  
Lin Jiang ◽  
Jingjing Zheng ◽  
Johnny S H Kwan ◽  
Sheng Dai ◽  
Cong Li ◽  
...  
Keyword(s):  
Negative Binomial ◽  
Negative Binomial Regression ◽  
Alternative Methods ◽  
Small Samples ◽  
Driver Mutations ◽  
Driver Genes ◽  
Cancer Driver ◽  
Binomial Regression ◽  
Technical Advances ◽  
Cancer Driver Genes

Abstract Genomic identification of driver mutations and genes in cancer cells are critical for precision medicine. Due to difficulty in modelling distribution of background mutation counts, existing statistical methods are often underpowered to discriminate cancer-driver genes from passenger genes. Here we propose a novel statistical approach, weighted iterative zero-truncated negative-binomial regression (WITER, http://grass.cgs.hku.hk/limx/witer or KGGSeq,http://grass.cgs.hku.hk/limx/kggseq/), to detect cancer-driver genes showing an excess of somatic mutations. By fitting the distribution of background mutation counts properly, this approach works well even in small or moderate samples. Compared to alternative methods, it detected more significant and cancer-consensus genes in most tested cancers. Applying this approach, we estimated 229 driver genes in 26 different types of cancers. In silico validation confirmed 78% of predicted genes as likely known drivers and many other genes as very likely new drivers for corresponding cancers. The technical advances of WITER enable the detection of driver genes in TCGA datasets as small as 30 subjects and rescue of more genes missed by alternative tools in moderate or small samples.

scholarly journals Foreign Aid Allocation from a Network Perspective: The Effect of Global Ties

2016 ◽  
Author(s):  
Liam Swiss
Keyword(s):  
Foreign Aid ◽  
Fixed Effects ◽  
Negative Binomial ◽  
Recipient Country ◽  
Negative Binomial Regression ◽  
World Society ◽  
Aid Allocation ◽  
Bilateral Aid ◽  
New Approach ◽  
Binomial Regression

This article examines competing explanations for foreign aid allocation on the global level and argues for a new approach to understanding aid from an institutionalist perspective. Using network data on all official bilateral aid relationships between countries in the period from 1975 through 2006 and data on recipient country ties to world society, the article offers an alternative explanation for the allocation of global foreign aid. Fixed effects negative binomial regression models on a panel sample of 117 developing countries reveal that global ties to world society in the form of non-governmental memberships and treaty ratifications are strong determinants of the network centrality of recipient countries in the global foreign aid network. Countries with a higher level of adherence and connection to world society norms and organizations are shown to be the beneficiaries of an increased number of aid relationships with wealthy donor countries. The findings also suggest that prior explanations of aid allocation grounded in altruist or realist motivations are insufficient to account for the patterns of aid allocation seen globally in recent years.

Cases and methods

2021 ◽  
pp. 43-71
Author(s):  
Emiliano Grossman ◽  
Isabelle Guinaudeau
Keyword(s):  
Fixed Effects ◽  
Negative Binomial ◽  
Negative Binomial Regression ◽  
Party Systems ◽  
Data Availability ◽  
Specific Structure ◽  
Far Right ◽  
Contemporary Politics ◽  
Binomial Regression ◽  
Legislative Agenda

This chapter presents the five cases. They are all wealthy Western European democracies. They feature quite a few similarities, concerning the evolution of turnout, electoral volatility, or the growing success of far-right parties. Yet, they also allow for some variation regarding their electoral systems, ranging from very majoritarian to very proportional and a resulting variety of party systems. Each country’s institutional setup and contemporary politics is briefly presented in turn. For each of the five countries, the chapter briefly discusses data availability and sources. For the five countries, we present the legislative agenda and the party agenda from the early 1980s to the late 2000s at least. Finally, we explain the choice of fixed-effects negative binomial regression, given the specific structure of the data and the goals of the study.

scholarly journals Analysis of deaths in a Nigerian hospital: the case of Federal Medical Centre (FMC), Umuahia

2015 ◽  
Vol 3 (1) ◽  
pp. 64 ◽  
Author(s):  
Emmanuel Ekpenyong ◽  
Iheanyi Emenike
Keyword(s):  
Poisson Regression ◽  
Causes Of Death ◽  
Negative Binomial ◽  
Medical Center ◽  
Circulatory System ◽  
Negative Binomial Regression ◽  
Binomial Regression ◽  
Level Of Significance ◽  
Causes Of Deaths ◽  
Over Dispersion

<p>A Poisson regression model was fitted to the causes of death in Federal Medical Center (FMC), Umuahia. The basic violation of the model which is over-dispersion was confirmed absent with the use of a plot of residual against fitted values, the Pearson statistics and the comparison of the log-likelihoods of the Poisson regression and Negative Binomial regression models. Diseases affecting the excretory and neurological systems respectively were found to be insignificant in the model, while the rest of the other causes of deaths – were found to be significant in the model at 5% level of significance. The leading causes of death in the hospital were determined to be diseases affecting the digestive and circulatory system, injuries, endocrine and neonatal related diseases, and cancers.</p>

Homeland Violence and Diaspora Insecurity: An Analysis of Israel and American Jewry

2019 ◽  
Vol 13 (1) ◽  
pp. 1-27
Author(s):  
Ayal K. Feinberg
Keyword(s):  
United States ◽  
Fixed Effects ◽  
Hate Crime ◽  
Negative Binomial ◽  
Negative Binomial Regression ◽  
The United States ◽  
Hate Crimes ◽  
Federal Bureau Of Investigation ◽  
American Jewry ◽  
Binomial Regression

AbstractJews and Jewish institutions have suffered the majority of reported religion-motivated hate crimes in the United States for nearly two decades. According to the Federal Bureau of Investigation (FBI), in 2014 the 609 reported anti-Semitic incidents made up 59% of all religious bias hate crimes alone. Rates of reported anti-Semitic hate crimes vary considerably over the course of a year. Yet, little scholarly attention has been given to what factors cause reported anti-Semitic hate crimes to fluctuate so substantially in the United States. This paper hypothesizes that violent Israeli military engagements are critical in explaining weekly surges of reported anti-Semitic hate crimes. Utilizing FBI hate crime data from 2001 to 2014 and fixed effects negative binomial regression models, consistent findings underscore that violent Israeli military engagements significantly increase the likelihood of a state reporting anti-Semitic hate crime. Most dramatically, their occurrence increases the likelihood of reported hate crime intimidating individuals or characterized as violent by nearly 35%. This paper underscores that homeland perpetrated violence can directly impact the security of diaspora communities.

scholarly journals KAJIAN EFEK SPASIAL KASUS DIFTERI DENGAN GEOGRAPHICALLY WEIGHTED NEGATIVE BINOMIAL REGRESSION (GWNBR)

2019 ◽  
Vol 3 (1) ◽  
pp. 91-104
Author(s):  
Diva Arum Mustika ◽  
Rani Nooraeni ◽  
Indonesian Journal of Statistics and Its Applications IJSA
Keyword(s):  
Negative Binomial ◽  
Immunization Coverage ◽  
Negative Binomial Regression ◽  
Alternative Methods ◽  
Spatial Effects ◽  
Main Method ◽  
Spatial Influence ◽  
Observation Area ◽  
Binomial Regression ◽  
Vaccine Availability

Diphtheria is an infectious disease caused by the Corynebacterium diphtheriae bacteria. Indonesia is the country with the most cases of diphtheria in Southeast Asia and ranks third in the world. In 2016, cases of diphtheria increased by 65 percent and became Extraordinary Events (KLB) in Indonesia, even though during 2013 to 2015 the number of cases of diphtheria has decreased. The province that has the highest number of diphtheria cases in Indonesia in 2016 is East Java. Diphtheria is centered and spread in certain districts / cities in East Java Province so that there are indications of spatial effects in the spread of diphtheria. Because data on the number of diphtheria cases overdispersed and indicated spatial effects in its spread, the main method used in this study was Geographically Weighted Negative Binomial Regression (GWNBR). This method will be compared with other alternative methods namely Poisson regression method and Negative Binomial Regression to get the best modeling. Based on the AIC value of each model it can be concluded that the best method for modeling the number of diphtheria cases is GWNBR. The modeling results with GWNBR show that there is indeed a spatial influence on the number of diphtheria cases and risk factors in East Java Province in 2016. The percentage of DPT-HB3 / DPT-HB-Hib3 immunization coverage is not significant in all observation areas, while the percentage of drug and vaccine availability is significant at entire observation area.

Sign in / Sign up

Export Citation Format

Share Document