/ Seminario de Estadística


Forecasting High-Dimensional Time Series via One-Sided Dynamic Principal Components
Ezequiel Smucler
Departamento de Computación, FCEyN-UBA

We define one-sided dynamic principal components (ODPC) for time series as linear combinations of the present and past values of the series that minimize the reconstruction mean squared error. It is shown that the ODPC can be successfully used for forecasting high-dimensional multiple time series. An alternating least squares algorithm to compute the proposed ODPC is presented. We prove that for stationary and ergodic time series the estimated values converge to their population analogues. We also that prove that asymptotically, when both the number of series and the sample size go to infinity, if the data follows a dynamic factor model, the reconstruction obtained with ODPC converges, in mean squared error, to the common part of the factor model. Monte Carlo results shows that forecasts obtained by the ODPC compare favorably with other forecasting methods based on dynamic factor models.

Lunes 5 de junio, 12Hs
Aula 1, Instituto de Cálculo - FCEyN


Fecha Expositor Filiación Título
05/06/2017Ezequiel SmuclerFCEyN-UBAForecasting High-Dimensional Time Series via One-Sided Dynamic Principal Components
19/05/2017Manuel Febrero-Bande Universidad de Santiago de Compostela Selección de variables en modelos de regresión aditivos funcionales
15/05/2017Mariela SuedFCEyN-UBAEstimadores robustos doblemente protegidos
10/04/2017Matthieu JonckherreFCEyN-UBAThe miracle of insensitivity for loss networks and the critical regime for load balancing in servers farms
03/04/2017Ruben ZamarUniversity of British ColumbiaRegression Phalanxes

Intendente Güiraldes 2160
Ciudad Universitaria
Pabellón II - 2do. piso
(C1428EGA) Buenos Aires
Teléfono directo/ Fax:
(54)(11) 4576-3375
(54)(11) 4576-3300 al 3309
interno 259