A comparison between classical (ARIMA) forecasting methods for time series and regression based on Support Vector Machines. A huge set of economic time series is available to train and validate foreasting methods
Skills required: basic computer science skills; python might be the language of chooice
Image credits: http://www.ece.umn.edu/users/cherkass/ee4389/SVR.html
Candidate: Alberto Paruta. Start: may 2016
Forecasting time series with Support Vector Regression