SSRS SVP Aniruddha Banerjee, Ph.D., Publishes Two New Papers

View the Latest Research from our SVP of Advanced Analytics

A Contribution on the Nature and Treatment of Missing Data in Large Market Surveys

Nonresponse (or missing data) is routinely encountered in large-scale surveys, and analysts often respond with simple case deletion or imputation even when the form of the “missingness” is unknown. As a result, subsequent statistical work can lead to misleading conclusions. This study first proposes a simple sequential method to attempt to identify the form of missingness, namely “missing completely at random” (MCAR), “missing at random” (MAR), and “not missing at random” (NMAR). Then, it assesses the effectiveness of the tests by experimentally generating nine data sets with missing data under the three forms of missingness.  An application to mobile commerce adoption shows that case deletion is acceptable when the form of missingness is MCAR, less so when it is MAR, and not at all when it is NMAR.

Read the Paper >>

E-Commerce Transactions, the Installed Base of Credit Cards, and the Potential Mobile E-Commerce Adoption

Mobile e-commerce (m-commerce) relaxes consumers’ temporal and geographic purchasing constraints and encourage the establishment of multichannel markets. It is often argued that rapid increase in smartphone penetration is the primary driver of m-commerce adoption, whereas others contend that early adoption of m-commerce is mostly by “relatively heavy” Internet commerce users. This article explores the strength of these influences within a nested multiple-service econometric framework, which allows for interdependency between m-commerce, e-commerce services, and the installed base of credit cards. The results reveal a complex situation in which credit cards facilitate e-commerce services, whereas m-commerce adoptions are driven by both prior e-commerce and online transaction activities. Also, higher incomes are negatively associated with proposed m-commerce adoption and, surprisingly, privacy concerns do not affect such adoption (except among the elderly).

Read the Paper >>