Prerequisites for travel modeling
Transportation planners and modelers come from diverse backgrounds. No definitive survey of their academic backgrounds is known to exist, much less agreement about an ideal slate of courses. Many modelers are thought to have graduate degrees in civil engineering with coursework in travel demand forecasting. Others have degrees in urban planning, economics, or related field. Those without an engineering background sometimes struggle with the quantitative foundations of advanced models, and seek references that will expand their grasp of "prerequisites for travel modeling." Some of the foundational areas include:
- Mathematics: A strong mathematical background is necessary for approaching the literature in travel modeling, although arguably less so for simply applying less complex models. Online calculus courses through Coursera, MIT and Wolfram Research are well reviewed. Those with weaker backgrounds might benefit from self-study of Jeff Gill's Essential mathematics for political and social research.
- Statistics: The number of online courses slanted towards statistics in data science has exploded over the past year. There are a large number of them available through EdX, with their two-part "Foundations of Data Analysis" being highly popular. There are several reviews of such programs online that can help hone the search. For those with little or no statistical background How to think about statistics can help begin the learning process, followed by Practical statistics for data scientists.
- Economics: The economics of transportation systems: a reference for practitioners is an outstanding review of essential concepts in this area.
- Machine learning: Not considered relevant as recently as two years ago this topic has rapidly become a hot topic in travel modeling. One of the most popular courses online is Stanford University's Machine learning, available through Coursera. Two books -- Learning from data and Introduction to statistical learning -- are approachable by those without strong mathematical backgrounds. Elements of statistical learning offers a deeper look into the topic.
The books and courses recommended by leading educators and practitioners in this field are changing quickly. Please leave comments below on great resources you come across.