A consistent estimator for the binomial distribution in the presence of “incidental parameters”: an application to patent data

Machado, Matilde P.

doi:10.1016/s0304-4076(03)00156-8

Cited by 10 publications

(26 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, Chamberlain (1985) provides results for some duration mod-els; Manski (1987) discusses the nonparametric binary choice model; Honoré (1992) provides a solution for censored regression; Kyriazidou (1997) for sample selection; Machado (2004) for binomial regression; Chamberlain (2010) for the binary choice logistic model; Magnac (2004) for a generalization of the conditional logit model; Shi et al (2016) for multinomial choice models; and Muris (2017) for logit ordered choice. For reviews of the literature, see Chamberlain (1984) (Section 3), Honoré (1992), Arellano and Honoré (2001) (Section 4), Arellano (2003), and Arellano and Bonhomme (2012) (Sections 2 and 4).…”

Section: The Incidental Parameter Problemmentioning

confidence: 99%

Binarization for panel models with fixed effects

Botosaru¹,

Muris²

2017

View full text Add to dashboard Cite

In nonlinear panel models with fixed effects and fixed-T , the incidental parameter problem poses identification difficulties for structural parameters and partial effects. Existing solutions are model-specific, likelihood-based, impose time homogeneity, or restrict the distribution of unobserved heterogeneity. We provide new identification results for the large class of Fixed Effects Linear Transformation (FELT) models with unknown, time-varying, weakly monotone transformation functions. Our results accommodate continuous and discrete outcomes and covariates, require only two time periods and no parametric distributional assumptions. First, we provide a systematic solution to the incidental parameter problem in FELT via binarization, which transforms FELT into many binary choice models. Second, we identify the distribution of counterfactual outcomes and a menu of time-varying partial effects. Third, we obtain new results for nonlinear difference-in-differences with discrete and censored outcomes, and for FELT with random coefficients. Finally, we propose rankand likelihood-based estimators that achieve √ n rate of convergence.

show abstract

Section: The Incidental Parameter Problemmentioning

confidence: 99%

Binarization for panel models with fixed effects

Botosaru¹,

Muris²

2017

View full text Add to dashboard Cite

show abstract

“…The literature shows several ways to access and enjoy the Big Data, usually in competitive intelligence literature, however, big data analysis processes are virtually countless, because depend on each database or combination of databases, the context and interest of analysis, continuity or perecividade data, of economic resources, technological and human disposition, described by Fleming et al (2001), Machado (2004) and Wu et al (2014) (Fleming & Sorenson, 2001;Machado, 2004; Xindong Wu, Xingquan Zhu Gong-Qing Wu, & Wei Ding, 2014). There are steps or common sub-processes in the data analysis process, described in the literature reviewed, which will be discussed briefly in the next section.…”

Section: Big Data Main Analysis Processmentioning

confidence: 99%

Conceptualization and theorization of the Big Data

Mazieri¹,

Soares

2016

Int. J. Innov.

View full text Add to dashboard Cite

The term Big Data is being used widely by companies and researchers who consider your relevant functionalities or applications to create value and business innovation. However, some questions arise about what is this phenomenon and, more precisely, how it occurs and under what conditions it can create value and innovation in business. In our view, the lack of depth related to the principles involved in Big Data and the very absence of a conceptual definition, made it difficult to answer these questions that have been the basis for our research. To answer these questions we did a bibliometric study and extensive literature review. The bibliometric studies were realized based in articles and citation of Web of Knowledge database. The main result of our research is the providing a conceptual definition for the term Big Data. Also, we propose which principles discovered can contribute with other researches that intend value creation by Big Data. Finally we propose see the value creation through Big Data using the Resource Based View as the main theory used for discuss that theme.

show abstract

“…Finally, we note that conditional likelihood estimators have also been proposed for other nonlinear fixed‐effects models with underlying exponential likelihoods. Examples of such models include the Poisson model (Hausman et al., ), the binomial regression model with logistic link function (Machado, ), and certain duration models (Chamberlain, ).…”

Section: Introductionmentioning

confidence: 99%

“…1 Finally, we note that conditional likelihood estimators have also been proposed for other nonlinear fixed-effects models with underlying exponential likelihoods. Examples of such models include the Poisson model (Hausman et al, 1984), the binomial regression model with logistic link function (Machado, 2004), and certain duration models (Chamberlain, 1985). For the semiparametric fixed-effects model with interval-censored outcomes, we propose a composite maximum-scoretype estimator to consistently estimate the slope parameters.…”

Section: Introductionmentioning

confidence: 99%

Interval censored regression with fixed effects

Abrevaya

Muris²

2019

J of Applied Econometrics

View full text Add to dashboard Cite

Summary This paper considers identification and estimation of a fixed‐effects model with an interval‐censored dependent variable. In each time period, the researcher observes the interval (with known endpoints) in which the dependent variable lies but not the value of the dependent variable itself. Two versions of the model are considered: a parametric model with logistic errors and a semiparametric model with errors having an unspecified distribution. In both cases, the error disturbances can be heteroskedastic over cross‐sectional units as long as they are stationary within a cross‐sectional unit; the semiparametric model also allows for serial correlation of the error disturbances. A conditional‐logit‐type composite likelihood estimator is proposed for the logistic fixed‐effects model, and a composite maximum‐score‐type estimator is proposed for the semiparametric model. In general, the scale of the coefficient parameters is identified by these estimators, meaning that the causal effects of interest are estimated directly in cases where the latent dependent variable is of primary interest (e.g., pure data‐coding situations). Monte Carlo simulations and an empirical application to birthweight outcomes illustrate the performance of the parametric estimator.

show abstract

A consistent estimator for the binomial distribution in the presence of “incidental parameters”: an application to patent data

Cited by 10 publications

References 18 publications

Binarization for panel models with fixed effects

Binarization for panel models with fixed effects

Conceptualization and theorization of the Big Data

Interval censored regression with fixed effects

Contact Info

Product

Resources

About