A better lemon squeezer? Maximum-likelihood regression with beta-distributed dependent variables.

By Michael Smithson, J. Verkuilen. Psychological methods, 2006

1,563 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/851779099a662b91c8153a9a40e069f0276eecbc

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

Beta Regression in R (2010, methodology): “We consider an application that analyzes reading accuracy data for nondyslexic and dyslexic Australian children (Smithson and Verkuilen 2006).”
Diagnostic tools in beta regression with varying dispersion (2011, methodology): “The application is based on reading accuracy data analysed by Smithson and Verkuilen (2006) from 44 individuals recruited from primary schools in the Australian Capital Territory.”
Alternative Estimating and Testing Empirical Strategies for Fractional Regression Models (2011, neutral): “…researcher may be interested in analyzing whether a variable contributes to the variance of  beyond its effect upon the mean, Paolino (2001) and also Smithson and Verkuilen (2006) assume  =  () and  = exp (), where  is a set of independent variables, potentially distinct from those…”
A Command for Fitting Mixture Regression Models for Bounded Dependent Variables Using the Beta Distribution (2018, methodology): “The standard beta regression model (Paolino 2001; Ferrari and Cribari-Neto 2004; Smithson and Verkuilen 2006) assumes that the dependent variable is continuous in the open unit interval (0, 1).”
On beta regression residuals (2008, methodology): “The next application is based on the data analysed by Smithson and Verkuilen 7 , obtained from Pammer and Kevan 6 .”
Mixed and Mixture Regression Models for Continuous Bounded Responses Using the Beta Distribution (2012, methodology): “Smithson and Verkuilen (2006) presented real-world examples of this and the present article includes additional illustrations.”
A quantile parametric mixed regression model for bounded response variables (2017, background): “similar to [32], to ease the interpretation of the coefficients.”
Demographics and Dynamics of Mechanical Turk Workers (2018, methodology): “Therefore, to examine if there is a correlation of the propensities with the demographic variables, we use a Beta regression [11, 31], which is explicitly designed to handle dependent variables that are Beta distributed and exhibit heteroskedasticity and skewness.”
Improved likelihood inference in beta regression (2011, methodology): “Applications of beta regression models can be found in Smithson and Verkuilen [6], Korhonen et al. [7], Silvestre et al. [8], Hubben et al. [9], among others.”
The Kumaraswamy distribution: median-dispersion re-parameterizations for regression modeling and simulation-based estimation (2012, methodology): “…that have been advanced for using the Beta distribution in regression models in a range of contexts and disciplines (Ferrari and Cribari-Neto 2004; Kieschnick and McCullough 2003; Paolino 2001; Smithson and Verkuilen 2006) apply unchanged if we substitute the Kumaraswamy for the Beta distribution.”
Beta regression for time series analysis of bounded data, with application to Canada Google Flu Trends. (2014, background): “1 (2001), Ferrari and Cribari-Neto (2004), and Smithson and Verkuilen (2006).”
Climate change in the Eastern Amazon: crop-pollinator and occurrence-restricted bees are potentially more affected (2020, methodology): “5)/number of observations] was applied to avoid zeros and ones (Smithson and Verkuilen 2006).”
A general class of zero-or-one inflated beta regression models (2011, neutral): “…models have been studied by Kieschnick & McCullough (2003), Ferrari & Cribari-Neto (2004), Espinheira, Ferrari & Cribari-Neto (2008a, 2008b), Paolino (2001), Smithson & Verkuilen (2006), Korhonen et al. (2007), Simas, BarretoSouza & Rocha (2010), and Ferrari & Pinheiro (2010), among others.”
A New Robust Regression Model for Proportions (2012, methodology): “In addition, as the parameters φi must be strictly positive, the log link that can be used for F−1 2 (·) is log(φi) = −wT i δ where we use the negative sign, as indicated by Smithson and Verkuilen (2006), to make the interpretation of the coefficients δ easier.”