Screening Covariates in Presence of Unbalanced Binary Dependent Variable

Author(s):  
Francesco Giordano ◽  
Marcella Niglio ◽  
Marialuisa Restaino
2012 ◽  
Vol 20 (4) ◽  
pp. 480-500 ◽  
Author(s):  
Justin Esarey ◽  
Andrew Pierce

In this article, we present a technique and critical test statistic for assessing the fit of a binary-dependent variable model (e.g., a logit or probit). We examine how closely a model's predicted probabilities match the observed frequency of events in the data set, and whether these deviations are systematic or merely noise. Our technique allows researchers to detect problems with a model's specification that obscure substantive understanding of the underlying data-generating process, such as missing interaction terms or unmodeled nonlinearities. We also show that these problems go undetected by the fit statistics most commonly used in political science.


1996 ◽  
Vol 2 (1) ◽  
pp. 1-11
Author(s):  
Norman E. Philp ◽  
Rumintha Wickramasekera

AbstractThe incessant trend towards the internationalisation of the marketplace will continue to dominate the agendas of managers of Australia's manufacturing establishments as they approach the next millennium. Empirical studies of the determinants of the firm's export marketing behaviour have been quite prolific and internationally comprehensive (Aaby and Slater 1989) and the characteristics and attitudes of the firm's main decision makers are often posed as important explanatory variables. The current study also examines the significance of managerial characteristics, commitment and attitudes towards exporting behaviour, by concentrating on a sample of firms drawn from a single industry (food and beverage processors) and with similar size and locational characteristics (small-medium firms located in regional Victoria).A logistic regression model with a binary dependent variable (ie current exporter vs non-exporter) and, initially, 19 independent variables was formulated and estimated. An optimal model utilising only six of the most significant, management-related variables was able to predict the probability of a firm being an exporter with an accuracy of over 90% . The importance of management's willingness to commit both its mind and its firm's resources to the export endeavour and its recognition of the importance of price competitiveness were significant discriminators of active export behaviour. Exporting managers were also found more likely to be tertiary educated and foreign language fluent but not necessarily any younger than their non-exporting counterparts.


2021 ◽  
Vol 29 (1) ◽  
Author(s):  
Hezlin Aryani Abd Rahman ◽  
Yap Bee Wah ◽  
Ong Seng Huat

Logistic regression is often used for the classification of a binary categorical dependent variable using various types of covariates (continuous or categorical). Imbalanced data will lead to biased parameter estimates and classification performance of the logistic regression model. Imbalanced data occurs when the number of cases in one category of the binary dependent variable is very much smaller than the other category. This simulation study investigates the effect of imbalanced data measured by imbalanced ratio on the parameter estimate of the binary logistic regression with a categorical covariate. Datasets were simulated with controlled different percentages of imbalance ratio (IR), from 1% to 50%, and for various sample sizes. The simulated datasets were then modeled using binary logistic regression. The bias in the estimates was measured using MSE (Mean Square Error). The simulation results provided evidence that the effect of imbalance ratio on the parameter estimate of the covariate decreased as sample size increased. The bias of the estimates depended on sample size whereby for sample size 100, 500, 1000 – 2000 and 2500 – 3500, the estimates were biased for IR below 30%, 10%, 5% and 2% respectively. Results also showed that parameter estimates were all biased at IR 1% for all sample size. An application using a real dataset supported the simulation results.


Author(s):  
Norman E. Philp ◽  
Rumintha Wickramasekera

AbstractThe incessant trend towards the internationalisation of the marketplace will continue to dominate the agendas of managers of Australia's manufacturing establishments as they approach the next millennium. Empirical studies of the determinants of the firm's export marketing behaviour have been quite prolific and internationally comprehensive (Aaby and Slater 1989) and the characteristics and attitudes of the firm's main decision makers are often posed as important explanatory variables. The current study also examines the significance of managerial characteristics, commitment and attitudes towards exporting behaviour, by concentrating on a sample of firms drawn from a single industry (food and beverage processors) and with similar size and locational characteristics (small-medium firms located in regional Victoria).A logistic regression model with a binary dependent variable (ie current exporter vs non-exporter) and, initially, 19 independent variables was formulated and estimated. An optimal model utilising only six of the most significant, management-related variables was able to predict the probability of a firm being an exporter with an accuracy of over 90% . The importance of management's willingness to commit both its mind and its firm's resources to the export endeavour and its recognition of the importance of price competitiveness were significant discriminators of active export behaviour. Exporting managers were also found more likely to be tertiary educated and foreign language fluent but not necessarily any younger than their non-exporting counterparts.


2015 ◽  
Vol 27 (6) ◽  
pp. 1181-1197 ◽  
Author(s):  
Alinda Kokkinou ◽  
David A. Cranage

Purpose – The purpose of the present study is to examine the effect of waiting lines on customers’ decisions between using a self-service alternative and using a service employee. As self-service technologies are expensive and time-consuming to design and implement, service providers need to understand what drives customers to use them. Service operators have the most control over waiting lines and flexibility in expanding capacity, either by adding service employees or by adding self-service kiosks. Design/methodology/approach – The study used online scenario-based surveys following a 4 (number of customers waiting for the self-service technology) × 4 (number of customers waiting for the service employee) design. A binary dependent variable was used to record participants’ choice of service delivery alternative. Findings – Using logistic regression, the authors found that customers are increasingly motivated to use self-service technology as the waiting line for the service employee grows longer. This effect is influenced by perceived usefulness, anticipated quality of the self-service technology, need for interaction and technology anxiety. Research limitations/implications – This study should be replicated in a real-world setting where actual behavior, and not only intention, can be measured. Practical implications – The study provides guidance on how service providers can design their service to take advantage of the motivating effect of waiting lines on usage of self-service technology. Originality/value – The present study is the first to combine a scenario-based experiment with a binary dependent variable to isolate the impact of waiting lines on the choice between using a self-service technology and using a service employee. The use of the binary dependent variable overcomes the ambiguity of extrapolating from a continuous measure of intention to draw conclusions about behavior, a binary variable.


Sign in / Sign up

Export Citation Format

Share Document