Special Issue of ASMBI for y‐BIS 2019: Recent Advances in Business Analytics and Data Science

Applied Stochastic Models for Business and Industries has recently published a special issue. The issue, Recent Advances in Business Analytics and Data Science, includes five papers that were presented at the 2019 y‐BIS (Young Business and Industrial Statisticians) Conference in Istanbul, Turkey hosted by the Mimar Sinan Fine Arts University. y‐BIS was created in 2008 as the group of young statisticians within the International Society for Business and Industrial Statistics (ISBIS). Its purpose is to bring together young researchers and professionals working on business, financial and industrial statistics, to help support their career development. A portion of the foreward to the issue, edited by Tahir Ekin, Gavino Puggioni and Paulo Canas Rodrigues, is included below: 

There are five papers in this special issue that covers a number of the business analytics and data science topics presented at the meeting. 

Castellani et al. investigate the use of machine learning methods for valuation of the solvency capital requirement of participating life insurance policies. In particular, they compare support vector regression and deep learning networks with the widely utilized least squares Monte Carlo techniques. They find machine learning methods to be promising alternatives in terms of accuracy and computational efficiency, which could help regulators and practitioners in life insurance industry.

Frigau et al. propose a novel network‐based semi‐supervised clustering approach associated with an outcome variable, that combines three phases: initialization, training, and agglomeration. The main objective of the proposed method is to find a partition of the original data into a certain number of disjoint clusters, that are the least possible overlapped with respect to the outcome. In that way, it is possible to order the clusters according to the mean values of the outcome while minimizing the overlap and maximizing the internal homogeneity and the external heterogeneity between clusters with respect to the outcome.

Artificial Neural Networks (ANN) have received a great deal of attention in the financial literature. Ilter et al. propose a novel feature selection procedure based on training the ANN with a hybrid approach based on Genetic Algorithms and Information Complexity Criterion. With a motivating application in credit scoring, their approach is illustrated with real benchmark datasets. While achieving robustness in selecting model complexity, the method shows improvement in classification accuracy when compared with standard training procedures.

The manuscript by Iorio and Pandolfo addresses the difficulty in defining a reasonable parametric model in multivariate quality control applications. Their approach, nonparametric in nature, tackles dimensionality reduction in high‐dimensional problems by using Lp data depth functions . The authors investigate their properties and behavior and show comparisons with other, more frequently used, depth functions.

Wang et al. explore the determinants of social promotion success, with a case study about crowdfunding projects. In particular, the authors use social influence, that is, reshare count and followers count, to evaluate the status of social promotions. Then, the social influence is modeled in a continuous framework in a way that the final influence of a project can be predicted progressively during the fundraising process. This enables monitoring the future influence during its whole duration, which allows project starters and fundraisers adjust the promotional strategies in advance.

ASMBI – Applied Stochastic Models in Business and Industry (formerly Applied Stochastic Models and Data Analysis) was first published in 1985, publishing contributions in the interface between stochastic modelling, data analysis and their applications in business, finance, insurance, management and production. In 2007 ASMBI became the official journal of the International Society for Business and Industrial Statistics