BEGIN:VCALENDAR
PRODID:-//Columba Systems Ltd//NONSGML CPNG/SpringViewer/ICal Output/3.3-
M3//EN
VERSION:2.0
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
DTSTAMP:20161108T095101Z
DTSTART:20161201T161500Z
DTEND:20161201T174500Z
SUMMARY:Degui Li (University of York)
UID:{http://www.columbasystems.com/customers/uom/gpp/eventid/}m79-itncbys
3-3bbfas
DESCRIPTION:Title: Semiparametric Ultra-High Dimensional Model Averaging
of Nonlinear Dynamic Time Series\n\nAbstract: We propose two semiparamet
ric model averaging schemes for nonlinear dynamic time series regression
models with a very large number of covariates including exogenous regre
ssors and auto-regressive lags\, aiming to obtain accurate forecasts of
a response variable by making use of a large number of conditioning vari
ables in a nonparametric way. In the first scheme\, we introduce a Kerne
l Sure Independence Screening (KSIS) technique to screen out the regress
ors whose marginal regression (or auto-regression) functions do not make
significant contribution to estimating the joint multivariate regressio
n function\; we then propose a semiparametric penalised method of Model
Averaging MArginal Regression (MAMAR) for the regressors and auto-regres
sors that survive the screening procedure\, to further select the regres
sors that have significant effects on estimating the multivariate regres
sion function and predicting the future values of the response variable.
In the second scheme\, we impose an approximate factor modelling struct
ure on the ultra-high dimensional exogenous regressors and use the princ
ipal component analysis to estimate the latent common factors\; we then
apply the penalised MAMAR method to select the estimated common factors
and the lags of the response variable that are significant. In each of t
he two semiparametric schemes\, we construct the optimal combination of
the significant marginal regression and auto-regression functions. Under
some regularity conditions\, we derive some asymptotic properties for t
hese two semiparametric schemes. Numerical studies including both simula
tion and an empirical application are provided to illustrate the propose
d methodology.
STATUS:TENTATIVE
TRANSP:TRANSPARENT
CLASS:PUBLIC
LOCATION:2nd Floor Boardroom\, Arthur Lewis Building\, Manchester
END:VEVENT
END:VCALENDAR