Faculty Mentor
Dr. Hari Adhikari
Abstract
Various events, such as changes in the interest rate or the hijacking of a commercial aircraft, can lead to significant shifts in airline stock performance. This study aimed to measure the impact of aviation-related news announcements on the stock performance of US airlines, focusing on different topics. The dataset included aviation news covering airlines, airports, regulations, safety, accidents, manufacturers, MRO, incidents, aviation training, general aviation, and others obtained from Aviation Voice. To uncover patterns that could explain the movements of US airline stocks, a natural language processing technique called Latent Dirichlet Allocation (LDA) was employed. The process involved text mining and topic modeling of the aviation-related data. The LDA model was then utilized to identify and capture specific topics mentioned in the news releases from Aviation Voice. By investigating the links between stock returns and the identified topics, the study revealed significant variations in financial performance across different topics. Notably, topics related to technology, fuel, and training positively impacted the short and long-term moving averages of US airline stocks. On the other hand, topics related to defense and travel costs only influenced the medium-term run. These findings shed light on the factors that influence US airline stock performance and provide valuable insights for investors and industry stakeholders.
Recommended Citation
Issoufou Anaroua, Amina
(2023)
"Analysis of US Airline Stocks Performance using Latent Dirichlet Allocation (LDA),"
Beyond: Undergraduate Research Journal: Vol. 7
, Article 3.
Available at:
https://commons.erau.edu/beyond/vol7/iss1/3
Included in
Business Analytics Commons, Business Intelligence Commons, Corporate Finance Commons, Finance and Financial Management Commons, Management Information Systems Commons, Science and Technology Studies Commons, Technology and Innovation Commons