Generalized extreme value regression for binary response data: An application to B2B electronic payments system adoption

TL;DR

This paper introduces a new generalized extreme value (GEV) based skewed link function for binary response regression, improving modeling flexibility and accuracy especially with imbalanced data, demonstrated through simulations and real-world electronic payments adoption data.

Contribution

The paper proposes a novel GEV-based skewed link function for binary regression, enhancing flexibility over existing models in handling imbalanced datasets.

Findings

GEV link models outperform traditional links in skewed data scenarios

The proposed model accurately predicts firm adoption probabilities

Application to real electronic payments data validates model effectiveness

Abstract

In the information system research, a question of particular interest is to interpret and to predict the probability of a firm to adopt a new technology such that market promotions are targeted to only those firms that were more likely to adopt the technology. Typically, there exists significant difference between the observed number of ``adopters'' and ``nonadopters,'' which is usually coded as binary response. A critical issue involved in modeling such binary response data is the appropriate choice of link functions in a regression model. In this paper we introduce a new flexible skewed link function for modeling binary response data based on the generalized extreme value (GEV) distribution. We show how the proposed GEV links provide more flexible and improved skewed link regression models than the existing skewed links, especially when dealing with imbalance between the observed number of 0's and 1's in a data. The flexibility of the proposed model is illustrated through simulated data sets and a billing data set of the electronic payments system adoption from a Fortune 100 company in 2005.