# The Impact of Covid-19 on Air Traffic: Spatiotemporal Forecasting with Deep Learning and Benchmarks 

## Description
Many global industries have been affected by the COVID 19 pandemic, the airline industry being one of the most heavily hit. As a result, flight trends around the world have drastically deviated from their normal patterns. This presents an interesting scenario to explore, especially with the transience and abnormality of effects of the Covid data. The phenomenon also created uncertainty for both passengers and airline companies, especially due to the multiple waves of virus mutations prompting the following questions: How should airlines plan future flights? When should passengers schedule their travels? In other words, given a country’s COVID situation, how should an airline/passengers plan ahead? To solve this problem, we present a solution in the form of an air traffic forecast using baseline models: AR and ARIMA, and deep learning models: LSTM (Long Short Term Memory), LSTM-mis-regression, LSTM-quantile-regression, and other models that capture the temporal complexity of such data. These algorithms are developed for time series analysis that are available to the open-source community, and can better serve airlines and passengers.