Topics Map > Services > Research Computing and Support > CCAST
Highlights - Financial, Economics, and Business Analytics
Principal Investigator: Limin Zhang (Accounting and Information Systems, North Dakota State University)
Dr. Zhang’s current research focuses on crawling and analyzing large amount of online financial documents and news articles. In particular, she is working with finance researchers at NDSU to investigate de-risking activities of U.S. companies regarding their defined-benefit pension plans.
With the support from CCAST at NDSU, the team collected and processed approximately 18 million documents (more than 54 TB total) in three months. After key sentences have been extracted from the documents, Dr. Zhang’s research group applied machine learning techniques to the dataset and performed text classification.
In another research project that requires sophisticated text and linguistic analysis of more than 800 GB of business news articles, Dr. Zhang uses CCAST’s HPC systems for storing and processing the data, taking advantage of the systems’ large storage, large memory, and parallel computing. As a result, processing time has been reduced from months to days.
Due to the large volume of data involved in such projects, the local HPC facility at NDSU has been critical to the success of the group’s research.
In addition, CCAST also plays an important role in the graduate education at NDSU. Students in the Master of Science in Business Analytics (MSBA) Program use CCAST systems to run big data and machine learning tools in their capstone projects and other course-related work, allowing them to learn and apply cutting-edge computing technologies to the analysis of real-world large datasets.
References
[1] J. Chen, R. Tian, and L. Zhang, “Strategic risk-shifting in corporate pension plans,” 2019 Financial Management Association International (IMA) Meeting, New Orleans, LA, USA.
[2] R. Tian, J. Chen, and L. Zhang, “De-risking DB pension or not? Facts from U.S. empirical data,” 2019 American Risk and Insurance Association (ARIA) Annual Meeting, San Francisco, CA, USA.
[3] L. Zhang, R. Tian, and J. Chen, “Text mining for pension de-risking analysis” (working paper).