A comparison of univariate probit and logit models using simulation

Abstract

Predictive analytics techniques are widely used in the application field, and the most common of these is fitting data with functions. The aim of function fittings is to predict the value of a response, by combing the regressors. Univariate probit and logit models are used for the same purposes when the response variable is binary. Both models used applied for the estimation of the functional relationship between response and regressors. The question of which model performs better comes to the mind. For this aim, a Monte Carlo simulation was performed to compare both the univariate probit and logit models under different conditions. In In this paper we considered the simulation of, employing latent variable approach with different sample sizes, cut points, and different correlations between response variable and regressors were taken into account. To make a comparison between univariate logit and probit models, Pearson residuals, deviations, Hosmer 186 Abeer H. Alsoruji et al. and Lemesshow, area under Receiver Operating Characteristic (ROC) curve, and Pseudo-R square statistics which are used for qualitative data analysis, were calculated and the results were interpreted.

Chronic patients as retirement-aged workers: the impact of employment-based health insurance and chronic conditions on health-related working capacity and late-life career participation

“…Given that the dependent variables (late-life career participation and health-related working capacity) were binary indicators, the discrete choice model was applied to examine the relationships of variables. The logit model with n regressors ( x 1 , x 2 ,---, x n ) performs better than the probit model with n regressors in the case of larger sample size, because when the sample size increases, the probability of observes in tail increases too (Alsoruji et al 2018 ). Thus, the logit model is used for the second regression (health-related working capacity) with the larger valid sample size.…”

Simulation-Based Assessment of Classification Methods: Statistical Models vs. Machine Learning Algorithms

“…In a similar but more recent study by Cakmakyapan and Goktas (2013) is that executed by Alsoruji et al (2018), the researchers conducted a simulation to compare the probit and logit models under various sample sizes, dependent-independent variables' correlation coefficients, and latent response In variable cut points. In the simulation, the regressand is influenced by three covariates from the standard multivariate normal distribution.…”