Thursday, March 23, 2017
Stats PhD Candidate Daniel Kowal’s work on hierarchical Gaussian process models was recently published in the Journal of Business & Economic Statistics. Authored along with Stats professors David Matteson and David Ruppert, “Functional Autoregression for Sparsely Sampled Data” proposes a model for forecasting and inference of functional time series data. Using theoretical results, a simulation study, and a financial application, the authors demonstrate the broad applicability and highly competitive forecasting capabilities of the model.