The 2019 Time Series & Forecasting Symposium is the third annual research meeting of the Time Series and Forecasting Research Group of the University of Sydney Business School.
This symposium aims to promote time series analysis and forecasting in business and other areas. The main themes of the event are, but not limited to, :
Dates: Monday 11 November and Tuesday 12 November 2019
Venue: ABS Case Study Lecture Theatre 2080 and 2090, Level 2, Abercrombie Building H70, The University of Sydney NSW 2006.
Cost: (Symposium rates include morning tea, afternoon tea and lunch)
Full Rate: $110.00
Student Rate: $55.00
Note: this does not include the Symposium dinner. You must register separately for the dinner.
Best Student Paper Competition: Student presentations will automatically be entered to this competition and a certificate and a prize will be awarded to the best paper.
Date: Monday 11 November 2019
Venue: The Little Snail, 3/50 Murray Street, Pyrmont NSW 2009 MAP
Cost: $50.00 per person (Including a 3-course dinner with one wine, beer or soft drink)
REGISTER FOR DINNER
Professor Morten Nielsen, Queen's University, Canada
Title: Inference on the dimension of the nonstationary subspace in functional time series
Abstract: This paper provides a statistical testing procedure to determine the number of stochastic trends of cointegrated functional time series taking values in the Hilbert space of square integrable functions defined on a compact interval. Our test is based on a variance ratio statistic, adapted to a possibly infinite dimensional setting. We derive its asymptotic null distribution and prove consistency of the test. Monte Carlo simulation results show good performance of our test and provide some evidence that it outperforms the existing testing procedure. The methodology is applied to three empirical examples: age-specific US employment rates, Australian minimum temperature curves, and hourly Ontario electricity demand.
Associate Professor Tomohiro Ando, University of Melbourne
Title: Quantile co-movement in stock markets with production linkages of firms: A spatial panel quantile model with unobserved heterogeneity
Abstract: This paper introduces a spatial panel quantile model with unobserved heterogeneity. The proposed model is capable of capturing high-dimensional cross-sectional dependence and allows heterogeneous regression coefficients. For estimating model parameters, a new estimation procedure is proposed. When both the time and cross-sectional dimensions of the panel go to infinity, the uniform consistency and the asymptotic normality of the estimated parameters are established. In order to determine the dimension of the interactive fixed effects, we propose a new information criterion. It is shown that the criterion asymptotically selects the true dimension. Monte Carlo simulations document the satisfactory performance of the proposed method. Finally, the method is applied to study the quantile co-movement structure of the U.S. stock market by taking into account the input-output linkages as firms are connected through the input-output production network.
The international keynote speaker was Professor Andrew J. Patton.
Download the 2018 program (pdf, 4.5MB)
The Sydney Time Series & Forecasting Symposium (TSF2017), held on Thursday 30 November 2017 and Friday 1 December 2017, was the inaugural annual research meeting of the Time Series and Forecasting Research Group of the University of Sydney Business School.
The international keynote speaker was Professor Javier Hidalgo, London School of Economics, UK.