Title: Stochastic Forecast Modeling and Evolutionary Estimation of Parameters Using Genetic Algorithm
Speaker: Dr. Shakshi Singhal
Date and Time: Tuesday, May 09, 2023, 10:00 am
Venue: IEOR 211, Seminar Room, 2nd floor IEOR Building
Abstract: Accurately predicting future outcomes based on historical data and other pertinent variables is essential for the industry’s decision-making processes. It has garnered significant attention in the information systems domain due to its importance in achieving strategic objectives and gaining a competitive advantage. Numerous quantitative forecast models have been utilized to anticipate the trajectory of diverse phenomena, viz. diffusion of innovations, malware propagation in cyber security, faults detection in software systems, new product dissemination in social networks, information propagation in tourism management, adoption of financial services, and disease transmission in epidemiology. The present research proposes stochastic differential equation-based forecast models for analyzing two processes: the adoption growth pattern of technological innovations under dynamic market size and the reliability growth behavior of software products under imperfect debugging. The interdisciplinary nature of predictive modeling is considered in this research and explores the application of forecast models in different areas. Two cases are examined wherein the stochastic parameter is modeled using exponential and Erlang distribution functions. Itô integral and Wiener process is employed to solve the stochastic differential equation. The empirical applications of the proposed methodology are verified using the real-life sales data of Samsung Galaxy and Apple iPhone Smartphones and failure data of the Open-Source Software projects, GNU Network Object Model Environment (GNOME), and Eclipse. A nature-inspired metaheuristic procedure, a genetic algorithm, is applied to estimate the model parameters. The predictive power of the developed models is compared with various benchmark studies. The data analysis is conducted using the R statistical computing software. The results demonstrate the proposed models’ efficacy in parameter estimation and predictive performance. A rolling cross-validation procedure is performed that shows the excellent forecasting capability of the developed models. The findings of the empirical analysis have provided supporting evidence of the stochastic increase in the market size of new technological products and fault generation in software systems. The numerical analysis results exemplify the proposed study’s capability of decision- making under uncertainty.
Speaker Bio: Dr. Shakshi Singhal is an Assistant Professor at the Fortune Institute of International Business (FIIB), New Delhi. Previously, she held the same position at the Jaypee Institute of Information Technology, Noida. She has also worked as a Research Assistant on an ICSSR-funded major research project related to e-waste management at the Delhi School of Management, Delhi Technological University, Delhi. Dr. Singhal obtained her Ph.D. in Operational Research from the University of Delhi. Besides this, she holds an M.Phil. and Master’s degree in Operational Research and a Bachelor’s degree in Computer Science from the University of Delhi. Dr. Singhal has taught several courses at the graduate and post-graduate levels, including Statistical Analysis & Decision-making, Logistics & Supply Chain Management, Quantitative Techniques, and Supply Chain Analytics, among others. She has also designed curriculums for these courses. Her teaching interest includes Statistics, Optimization Techniques, Software Reliability, Business Analytics, and Operations Management. Dr. Singhal’s research interest focuses on areas such as Predictive Modeling, Technology Diffusion, Optimization, Decision-making under uncertainty, Software Management, and Waste Management. Her research work has been published in several international journals, including Technological Forecasting & Social Change, Annals of Operations Research, Journal of Systems & Software, and Benchmarking: An International Journal. She is also a Doctoral Student Supervisor for the Executive Fellow Programme in Management (EFPM) and a lifetime member of the Society for Reliability Engineering, Quality, and Operations Management (SREQOM). At FIIB, she is involved in institution-building activities as the Research Head, coordinating and validating research activities such as publications, workshops, and funding under the guidance of the Research Chair. She is also the Head of AACSB Accreditation Standard 8, responsible for compiling and validating information, writing reports, and presenting to the Accreditation Chair. She also serves as an Editor for the International Journal of Systems Assurance Engineering and Management, Springer.