Objective

• A US-based sell-side analyst wanted to leverage data on airfares as a lead indicator and, thereby, improve existing fundamental research techniques to predict key trends in the US airline industry
• However, collection and analysis of airfare data was restricted to a small sample size, extremely timeconsuming and involved manual processing
  

Crisil Integral IQ solution

• The Crisil team identified 2,500 flight routes within the scheduled domestic US air-travel market, representing more than 80% of seating capacity
• The team wrote a web-scraping code in Python for automated fetching of the published airfares for business and leisure passenger segments from Google Flights, KAYAK and Cheapflights, among others, based on future travel date, selected origin and destination, number of stops, baggage options and travel class
• Future weighted-average airfares of each airline under research coverage were calculated based on an airline’s relative exposure to an airport and its domestic seat share
• Deep analysis of the patterns and statistics of the collected data was undertaken to predict near-term demand, capacity, pricing trends and profitability. The trajectory of future airfares was presented on week-over-week and year-over-year basis
• The signals drawn from the airfares data were baked into the analyst’s model estimates 

Client Impact

• Use of airfares data as a lead indicator significantly improved the analyst’s ability to predict key near-term industry trends and strengthened conviction on their views
• The dataset gets regular citations in the client’s research reports, and remains the most sought-after among buy-side clients
  

Questions

Looking for high-end research and risk services? Reach out to us at: