Grant recipients:
Yulia Gel, Jeanette O'Hara-Hines, Vyacheslav Lyubchich*, Department of Statistics and Actuarial Science
*Graduate Student Co-Applicant
(Completed. Project timeline: September 2013-August 2014)
 |
 |
 |
| Yulia Gel | Jeanette O'Hara-Hines | Vyacheslav Lyubchich |
Project Description
In a modern business environment, managers are increasingly required to perform decision-making and evaluate related risks based on quantitative information in the face of uncertainty, which in turn increases demand for business professionals with sound skills and hands-on experience with statistical data analysis. Computer-based training technologies allow the new cadre of business professionals to obtain such a hands-on experience in an environment where mistakes can be made and outcomes can be measured. In this project, we develop a new E-Learning tool designed to apply methodological forecasting concepts to real-life business and finance problems through an interactive self-learning and self-assessing module of online case studies and investigate its effectiveness to improve educational outcomes.
Questions Investigated
We developed three case studies involving short, medium, and long term prediction of sales of Australian fortified wine, 25 years of household financial obligations, and daily stock returns of a U.S. electronics company within the Learning Management System UW-ACE. The goal is to teach students that there is no ”perfect recipe” in business forecasting or modeling, and to show that the same data sets can be approached from different angles, depending on the analyst’s task.
Findings/Insights
We regress the students’ final exam grade on their case studies grade, including the average Grade Point Average (GPA) from previous terms, major, term of study, and gender as covariates in an attempt to control for the incoming intelligence of students. Similar to our earlier findings in 2011 where we used the Learning Management System Angel, in spring 2013 we observe that the case studies grade is significantly associated with the final exam grade, even after accounting for all other covariates. Hence, we can conclude that the proposed interactive E-Learning tool with interactive case studies provides a noticeable positive impact on educational outcomes in the higher level Forecasting course.
Dissemination and Impact
- At the individual level: The new E-Learning tool was shared with our instructors of STAT443 Forecasting.
- At the national and/or international levels: This research was presented at the following conferences:
1. X. Wang, Y. R. Gel, J. O’Hara Hines, V. Lyubchich (2014). Assessing Interactive E-Learning Techniques for Teaching Time Series and Forecasting. Student Conference at the Annual Meeting of the Canadian Statistical Society (SSC), Toronto, ON, May 2014.
2. Gel, Y.R., O'Hara-Hines, J., Chen, H., Noguchi, K., and Schoner, V. (2014). Developing and Assessing E-Learning Techniques for Teaching Forecasting. Journal of Education for Business, 89(5), 215--221.
3. The new tool was also discussed informally at the Joint Statistical Meeting (JSM) and the Canadian Meteorological and Oceanographic Society, with the intent to potentially expand the tool to other applied statistics courses for non-statistics majors.
References
References (pdf)