Past Meetup

Luncheon: Statistical Analysis of Online Travel: Purchase Timing

This Meetup is past

38 people went

Location image of event venue


All registrants must finalize registration here:

Statistical Analysis of Online Travel: Purchase Timing and Other Applications

In the Chicago area, statisticians have found a home for a number of years in the fields of financial services, market research, media, health care, and academia. Until fairly recently, "online travel" in general (or Orbitz in particular) wasn't even a blip on the Chicago area statistical radar. This talk will provide an insider view of what has happened in the last 5 years to take Orbitz from 1 to 20+ people using key concepts from the field of statistics on a daily basis. A broad survey of examples will be provided to illustrate how statistical analysis and related techniques (including operations research and machine learning) are influencing Web design, eMarketing, customer interactions, supplier interactions, pricing and promotions, and loyalty programs at Orbitz Worldwide. The example of purchase timing analysis will be explored in more detail. In particular, the application of decision tree techniques to purchase timing analysis will be discussed, along with the pros and cons of different techniques (CART vs CHAID) and some related words of caution for both purchase timing analysis and predictive analytics.

Tim Krick is a Senior Director of Advanced Analytics at the online travel agency Orbitz Worldwide, where he leads customer analytics initiatives for Orbitz and several affiliated brands around the world (including Cheaptickets, eBookers, and HotelClub). Tim has over 20 years of experience in data analytics and statistical analysis; he worked at Nielsen and Deloitte before joining Orbitz 5 years ago. His educational background includes an undergraduate degree in Mathematical Methods from Northwestern and a masters degree in Computer Science (focused on artificial intelligence and data mining) from DePaul. Tim often observes that while the names we give to our work have changed a bit over the years (e.g., statistics, data mining, advanced analytics, data analytics, predictive analytics, machine learning, Big Data), core ideas and approaches from the field of Statistics continue to be foundational.