A zero-inflated ordered probit model, with an application to modelling tobacco consumption

Harris, Mark N.; Zhao, Xueyan

doi:10.1016/j.jeconom.2007.01.002

Cited by 168 publications

(181 citation statements)

References 19 publications

Supporting

Mentioning

175

Contrasting

Unclassified

Order By: Relevance

“…Harris and Zhao (2007) show in Monte Carlo experiments that the maximum likelihood estimator of this model performs well in finite samples. We group the few price reductions of $5 or more together and similarly group the few price increases of $5 or more so that we only need to estimate thresholds down to C −5 τ and up to C 5 τ .…”

Section: Estimation Detailsmentioning

confidence: 93%

Costs of Managerial Attention and Activity as a Source of Sticky Prices: Structural Estimates from an Online Market

2016

View full text Add to dashboard Cite

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in

show abstract

Section: Estimation Detailsmentioning

confidence: 93%

Costs of Managerial Attention and Activity as a Source of Sticky Prices: Structural Estimates from an Online Market

2016

View full text Add to dashboard Cite

show abstract

“…Four independent variables X 1 to X 4 were generated from Uniform U (2, 5), Normal N (1, 1.5), Exponential (1), and Bernoulli distributions B(.3), accordingly. Although the covariates in the inflation parts and the regular model parts could be identical and the results would not be altered, to ensure exclusion restrictions and gain possible enhancement on precisions of parameter estimates, we varied the set of predictors by the outcomes (Bagozzi & Mukherjee, 2012;M. N. Harris & Zhao, 2007).…”

Section: Monte Carlo Experimentsmentioning

confidence: 99%

“…For other types of discrete outcomes, such as binary, multinomial or ordinal, various single value inflated models were developed, including: binary choice model with misclassification (Hausman, Abrevaya, & Scott-Morton, 1998), zero-inflated Bernoulli model (Diop, Diop, & Dupuy, 2016), zero-inflated binomial model (Hall, 2000;Vieira et al, 2000), zero-inflated ordered probit model (M. N. Harris & Zhao, 2007), baseline or zero inflated multinomial logit model (Bagozzi, 2016;Diallo, Diop, & Dupuy, 2017), and middle category inflated ordered model (Bagozzi & Mukherjee, 2012). Similar extension has been made to incorporate inflation other than zero for multinomial or ordinal outcomes (Sweeney, Haslett, & Parnell, 2014) Following Begum et al (2014), a further generalization could be made if the PMF is replaced by other discrete distributions, e.g., Binomial, Multinomial, Negative Binomial, etc.…”

Section: Introductionmentioning

confidence: 99%

Generalized Inflated Discrete Models: A Strategy to Work with Multimodal Discrete Distributions

Cai¹,

Xia²,

Zhou³

2017

Preprint

View full text Add to dashboard Cite

W o r k i n g P a p e rAnalysts of discrete data often face the challenge of managing the tendency of inflation on certain values. When treated improperly, such phenomenon may lead to biased estimates and incorrect inferences. This study extends the existing literature on single value inflated models, and develops a general framework to handle variables with more than one inflated value. To assess the performance of the proposed maximum likelihood estimator, we conducted Monte Carlo experiments under several scenarios for different levels of inflated probabilities under Multinomial, Ordinal, Poisson, and ZeroTruncated Poisson outcomes with covariates. We found that ignoring the inflations leads to substantial bias and poor inference of the inflations-not only for the intercept(s) of the inflated categories, but other coefficients as well. Specifically, higher values of inflated probabilities are associated with larger biases. By contrast, the Generalized Inflated Discrete Models (GIDM) perform well with unbiased estimates and satisfactory coverages even when the number of parameters that need to be estimated is quite large. We showed that model fit criteria, such as AIC, could be used in selecting the appropriate specifications of inflated models. Lastly, the GIDM was implemented using large-scale health survey data as a comparison to conventional modeling approaches such as various Poisson and Ordered Logit models. We showed that the GIDM fits the data better in generalThe current work provides a practical approach to analyzing multimodal data that exists in many fields, such as heaping in self-reported behavioral outcomes, inflated categories of indifference and neutral in attitude surveys, large amounts of zero, and low occurrences of delinquent behaviors.

show abstract

“…[ Following Harris and Zhao (2007), let us define a discrete random variable y that takes discrete ordered values of , and let denote a binary variable indicating the split between regime 0 (non-participants) and regime 1 (participants). The indicator r is related to a latent variable such that for and for .…”

Section: Introductionmentioning

confidence: 99%

“…Allowing the error terms from the first stage probit equation and the second stage ordered probit equation (i.e., and ) to be correlated, then, according to Harris andZhao (2007, p.1076), the full probabilities for are given by:…”

Section: Introductionmentioning

confidence: 99%

Video games playing: A substitute for cultural consumptions?

Borowiecki

Prieto‐Rodríguez

2014

J Cult Econ

View full text Add to dashboard Cite

This article provides an applied investigation of video game users. We estimate zero-inflated ordered probit models to control for an excess of zeros in our ordinal dependent variable. We find that video games playing is not negatively associated with the involvement in other cultural practices. On the contrary, instead of being a substitute for more traditional forms of cultural consumption, the probability of game playing increases with the consumption of other cultural goods (e.g., listening to music or watching television) or active involvement in artistic activities (e.g., writing or visual arts production). Game playing is in general an urban phenomenon, and it is positively associated with the ownership of home equipment and access to new technologies but decreases with a person's greater time restrictions. The main differences to the traditional art formats is that game playing particularly appeals to younger, usually male, cohorts.

show abstract

A zero-inflated ordered probit model, with an application to modelling tobacco consumption

Cited by 168 publications

References 19 publications

Costs of Managerial Attention and Activity as a Source of Sticky Prices: Structural Estimates from an Online Market

Costs of Managerial Attention and Activity as a Source of Sticky Prices: Structural Estimates from an Online Market

Generalized Inflated Discrete Models: A Strategy to Work with Multimodal Discrete Distributions

Video games playing: A substitute for cultural consumptions?

Contact Info

Product

Resources

About