Second Announcement of Time Series & Forecasting Symposium (TSF2018): 13-14/11/2018 Sydney

[Boris Choy]

 

Dear ANZECMET list members,

The Time Series & Forecasting Symposium 2018 (TSF2018) is an annual research event of the Time Series and Forecasting Research Group, the University of Sydney Business School. This symposium aims to promote time series analysis and forecasting in business and other areas. We welcome oral and poster presentations in all areas related to time series and forecasting, and especially encourage contributions in the main themes:  Bayesian time series, forecast combination, multi-step and long range forecasting, forecasting higher moments, volatility modelling and risk forecasting, robust inference for time series, copula time series modelling, dynamic copulas and other dependence measures. Registration is now open!

Dates: Tuesday, 13 November 2018 & Wednesday, 14 November 2018

Time:   9:00am – 5:00pm (Morning tea, lunch and afternoon tea are provided)

Venue: The Refectory and Room 5050, Level 5, Business School Abercrombie Building (H70), The University of Sydney Business School

             Map: Google search “The University of Sydney Business School”

International keynote speaker: Prof Andrew J. Patton, Duke University, UK (Please see title and abstract below)

Registration Fee:  $110 (incl. GST) for academic and industry participants and $55 (incl. GST) for full-time students. All registrations include morning tea, afternoon tea and lunch

Registration webpage: Please register at http://sydney.edu.au/business/research/tsfrg/events/tsf2018 before 9 November 2018

Abstract submission: Please send to  tsf.sy...@sydney.edu.au by email before 30 September 2018

Enquiries: tsf.sy...@sydney.edu.au

Local Organising Committee: Boris Choy and Henry Leung (Co-Chairs), Jennifer Chan, Richard Gerlach, Simon Kwok, Artem Prokhorov, Chao Wang 

 

Best regards,

Local Organising Committee, TSF2018

 

International Keynote Speaker:           

Professor Andrew J. Patton, Duke University, UK.

'Estimation and Inference for Large Panel Copula Models' 

 
Abstract:
We consider methods for modelling the copula of very large collections of random variables. Our asymptotic analysis is based on allowing both the number of variables and the length of each sample to diverge to infinity. We consider a class of factor copula models, with parameters estimated using the simulated method of moments type approach presented in Oh and Patton (2013, JASA). We provide conditions under which the estimated parameters are consistent and asymptotically normal. The factor loadings are assumed to be drawn randomly from a distribution whose parameters we estimate along with the other parameters of the model. We consider some measures of market-wide risk and show how to conduct inference on such measures. We apply the methods to monthly returns on 4,500 U.S. firms over the period 1926 to 2016.