Using Heterogeneous Choice Models to Compare Logit and Probit Coefficients Across Groups

By Richard Williams. 2009

513 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/e4fba58afc0f15ac856634ad6df851341218ef5a

How researchers see this paper

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

In Economics and Econometrics

In Statistics and Probability

Citations in the corpus

Using Heteroskedastic Ordered Probit Models to Recover Moments of Continuous Test Score Distributions From Coarsened Data (2017, methodology): “The ordered probit model is sometimes referred to as an ordered choice model (Williams, 2009) or as a location-scale model (Cox, 1995; McCullagh, 1980).”
Adverse Childhood Experiences and Adolescent Delinquency in a High-Risk Sample (2018, background): “Allison (1999) and others (Karlson, Holm, & Breen, 2012; Mood, 2009; Williams, 2009) have proposed a variety of methods for testing group differences when modeling binary outcomes, but each has been subject to criticisms, and there is currently no consensus regarding the optimal method for testing…”
Logistic Regression: Why We Cannot Do What We Think We Can Do, and What We Can Do About It (2010, methodology): “Such models can be estimated in common statistical software, such as Stata [using the oglm command (Williams 2006b)], SPSS and Limdep, and are based on the idea of estimating one ‘choice equation’ that models the effect of a range of variables on the outcome, and one ‘variance equation’ that models…”
The Family's Financial Support as a “Poisoned Gift”: A Family Embeddedness Perspective on Entrepreneurial Intentions (2017, result): “Although we need to be careful when comparing OR with that reported in studies which use different data sets (Williams 2009), our OR and thus effect sizes could be regarded as satisfactory compared with other studies that test models with direct (Kam et al.”
Do protest responses to a contingent valuation question and a choice experiment differ? (2008, methodology): “The minimum in PROTEST leads to a Pr(WTP=1) of 96% and a maximum to a probability of 1%. 8 Additionally, we applied more sophisticated tests to compare logit coefficients across groups (Allison 1999; Williams 2006 for an improvement on Allison’s test).”
Stock Market Literacy, Trust, and Participation (2015, unknown): “…di⁄erent predictor variables in the various probit model speci(cid:133)cations, the scaling of the response variable changes and therefore the changes in estimated coe¢ cients might not entirely be due to the suppressor e⁄ect (see Winship and Mare, 1984; Williams, 2009; Mood, 2010; among others).”
Understanding and interpreting generalized ordered logit models (2016, background): “For example, Williams (2009, 2010) notes that the ordered logit model also assumes that errors are homoscedastic, and that differences across groups in residual variability may cause tests of the proportional odds assumption to fail.”
Ordinal regression: A review and a taxonomy of models (2021, methodology): “Allison's article stimulated extensive research that aimed at avoiding such misleading results, see, for example, Williams (2009), Mood (2010), Karlson et al. (2012), Breen et al. (2014), Rohwer (2015), and Tutz (2019).”
On Group Comparisons With Logistic Regression Models (2020, background): “Data from Google scholar show that by February 28, 2017, Allison (1999), the earliest key contribution, had received 742 citations, Williams (2009) 262, and Mood (2010) a remarkable 1,053.”
The statistical analysis of highway crash-injury severities: a review and assessment of methodological alternatives. (2011, methodology): “Xie et al. (2009) used Markov chain Monte Carlo algorithms and found the Bayesian models to produce more reasonable parameter estimates and better predictive capabilities for smaller sample sizes.”
Using Predictions and Marginal Effects to Compare Groups in Regression Models for Binary Outcomes (2018, methodology): “Allison (1999) and Williams (2009) developed tests for comparing regression coefficients that account for differences in unobserved heterogeneity.”