Predicting Human Mobility in Urban Settings. The case of Manhattan, NY

AbstractIn this study we are building a prediction model for human mobility based on GPS-records collected in New York City in June 2016. We are building a prediction model based on a Markov Chain and including contextual variable at the neighborhood level. Our assumption is that contextual variable have a impact on mobility patterns. Through hypotheses tests we can confirm our hypotheses and we observe that indeed contextual variable impact the accuracy of the prediction model. We can therefore infere that contextual variables influence mobility behavior. To depict the influence of single variables on human mobility further research is necessary.