2015
DOI: 10.2139/ssrn.2642854
|View full text |Cite
|
Sign up to set email alerts
|

Applications of Fractional Response Model to the Study of Bounded Dependent Variables in Accounting Research

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
23
0

Year Published

2016
2016
2024
2024

Publication Types

Select...
7
2
1

Relationship

0
10

Authors

Journals

citations
Cited by 41 publications
(23 citation statements)
references
References 71 publications
0
23
0
Order By: Relevance
“…First, a descriptive analysis of participants’ characteristics and relevant variables was conducted. Second, we conducted three fractional probit response models (Papke & Wooldridge, )—an extension of the general linear model (Gallani, Wooldridge, & Krishnan, ). Using fractional response modelling guarantees that all predicted values lie in the unit interval between 0 and 1.…”
Section: Methodsmentioning
confidence: 99%
“…First, a descriptive analysis of participants’ characteristics and relevant variables was conducted. Second, we conducted three fractional probit response models (Papke & Wooldridge, )—an extension of the general linear model (Gallani, Wooldridge, & Krishnan, ). Using fractional response modelling guarantees that all predicted values lie in the unit interval between 0 and 1.…”
Section: Methodsmentioning
confidence: 99%
“…Proportional variables can only take on values equal to or between zero and one, meaning that it is unlikely that the effect of an independent variable on the dependent variable remains the same across values of x (Papke and Wooldridge, 1996). Fractional models are designed to accommodate these peculiarities, including non-linearity and a large amount of observations at the extremes (Gallani et al., 2015: 5–6). To simplify the interpretation of coefficients, I calculate average marginal effects for all variables.…”
Section: Empirical Analysismentioning
confidence: 99%
“…FRM is able to handle the nature of the dependent variable (DEA scores) that takes the value inside the interval (0,1), regardless of the availability of observed frontier values. According to Ramalho et al (2010) and Gallani et al (2015), FRM is the most advantageous model for continuous data with values bounded from 0 to 1. Ramalho et al (2016) propose a general regression model regarding the fractional nature of response variables as equation 1:…”
Section: Non-parametric and Parametric Research Approachesmentioning
confidence: 99%