VAR Models and Mixed Frequency Data
Professor Manfred Deistler (Vienna University of Technology)
SYSTEMS AND CONTROL SERIESDATE: 2013-11-14
TIME: 11:00:00 - 12:00:00
LOCATION: RSISE Seminar Room, ground floor, building 115, cnr. North and Daley Roads, ANU
CONTACT: JavaScript must be enabled to display this email address.
ABSTRACT:
This lecture is concerned with identifiability of an underlying high frequency multivariate AR system from mixed frequency observations. The main results show that on a generic subset of the paramter space identifiability holds. We deal with AR systems with nonsingular and singular innovation variance matrices, the latter being important for dynamic factor models. Such models, which are used for high dimensional time series, and their properties will be discussed. We analyze the case of stock and flow variables. If we have identifiability, the parameters and thus all second moments of the output process at the high sampling frequency can be estimated consistently from mixed frequency data. Then linear least squares methods for forecasting, nowcasting and interpolation of nonobserved output variables can be applied.
BIO:
Manfred Deistler is an Emeritus Professor of Econometrics and System Theory at Vienna University of Technology. He received his Dr. techn. (approximately corresponding to a PhD) from Vienna University of Technology in 1970. Manfred Deistler has served on the editorial board of a number of journals, at present he is an Associate Editor of Journal of Econometrics and he is a member of the Advisory Board of Econometric Theory. He is a Fellow of the Econometric Society, a Fellow of IEEE and of Journal of Econometrics. Manfred Deistler Is research interests are in econometrics, system identification and time series analysis. As far as theory and methods are concerned the focus of his work is on structure theory and estimation for multivariate ARMAX- and state space systems and for linear dynamic factor- and errors- in- variables models. His current research interests are modeling of high-dimensional time series, in particular by generalized dynamic factor models, singular AR systems (structure theory and Yule Walker estimators) and time series modeling from mixed frequency data (generic identifiability, asymptotic distributions of extended Yule Walker estimators, blocking). As far as applications are concerned, his current interests are: Analysis of electroencephalograms (focus detection in epileptic seizures based on ECoGas, EEG and Alzheimer) and forecasting of financial assets.
