Analysis of Us Airline Stock Performance Using Latent Dirichlet Allocation (LDA)

Tracking the news of airlines, accidents, MRO, and other related aviation topics, we measure the impact of such news on the stock performance of US airlines. We use Latent Dirichlet Allocation (LDA) to search for patterns that can explain the movements of US airline stocks. First, we mine the data through text mining and topic modeling. Second, we employ LDA to identify and capture topics from news releases. Finally, we investigate the links between stock returns and the topics identified. Our findings provide evidence that financial performance varies significantly across topics. Topics related to technology, fuel, and training positively affect US airline stocks in the short and long-term moving averages. In contrast, defense and travel cost-related topics only affect the medium-term run. Our paper contributes to a better understanding of the effect of the news on one of the most celebrated industries