The SEED Team Online broadcasting at https://webconf.vc.dfn.de/optimization/. For more details kindly see Instructions.

 

 

Speaker Title Date and Time Host
Hong Kong, Singapore and Taiwan (UTC +8) Japan and Korea
(UTC +9)
Germany
(UTC +1)
U.K.
(UTC)
Hideitsu Hino Current Dipole Localization from EEG with Birth-Death Process Wed, 9 Jan 2019
16:00 - 17:00
Wed, 9 Jan 2019
17:00 - 18:00
Wed, 9 Jan 2019
09:00 - 10:00
Wed, 9 Jan 2019
08:00 - 09:00
ISM
Guido Germano Integral transform methods and spectral filters for the pricing of exotic options Fri, 14 Dec 2018
17:30 - 18:30
Fri, 14 Dec 2018
18:30 - 19:30
Fri, 14 Dec 2018
10:30 - 11:30
Fri, 14 Dec 2018
09:30 - 10:30
UCL
Multiple Speakers
Workshop on Social Media Analytics Thu, 29 Nov 2018
14:40 - 16:20
Thu, 29 Nov 2018
15:40 - 17:20
Thu, 29 Nov 2018
07:40 - 09:20
Thu, 29 Nov 2018
06:40 - 08:20
HSU
Simone Righi Social closure and the evolution of cooperation via indirect reciprocity Fri, 16 Nov 2018
18:00 - 19:00
Fri, 16 Nov 2018
19:00 - 20:00
Fri, 16 Nov 2018
11:00 - 12:00
Fri, 16 Nov 2018
10:00 - 11:00
UCL
Shuhei Mano A Direct Sampler from Log-affine Models with Aid of Computational Algebra Wed, 31 Oct 2018
16:00 - 17:00
Wed, 31 Oct 2018
17:00 - 18:00
Wed, 31 Oct 2018
09:00 - 10:00
Wed, 31 Oct 2018
08:00 - 09:00
ISM
Tomaso Aste Blockchain for a Complex Society Fri, 19 Oct 2018
15:00 - 16:00
Fri, 19 Oct 2018
16:00 - 17:00
Fri, 19 Oct 2018
08:00 - 09:00
Fri, 19 Oct 2018
07:00 - 08:00
NUS
Giacomo Livan Reciprocity and success in academic careers Thu, 11 Oct 2018
17:00 - 18:00
Thu, 11 Oct 2018
18:00 - 19:00
Thu, 11 Oct 2018
11:00 - 12:00
Thu, 11 Oct 2018
10:00 - 11:00
UCL
Richard Peterson Sentimental Markets:  How Information Flow Drives Patterns in Asset Pricing Thu, 11 Oct 2018
15:00 - 16:00
Thu, 11 Oct 2018
16:00 - 17:00
Thu, 11 Oct 2018
9:00 - 10:00
Thu, 11 Oct 2018
8:00 - 9:00
NUS
Multiple Speakers Workshop on Risk Analytics Thu, 4 Oct 2018
18:30 - 20:30
Thu, 4 Oct 2018
19:30 - 21:30
Thu, 4 Oct 2018
12:30 - 14:30
Thu, 4 Oct 2018
11:30 - 13:30
HKUST
Ostap Okhrin Flexible HAR Model for Realized Volatility Fri, 28 Sep 2018
16:30 - 17:30
Fri, 28 Sep 2018
17:30 - 18:30
Fri, 28 Sep 2018
10:30 - 11:30
Fri, 28 Sep 2018
09:30 - 10:30
NUS
Xuan Leng Bias correction for the maximum likelihood estimator of the extreme value index Fri, 21 Sep 2018
16:30 - 17:30
Fri, 21 Sep 2018
17:30 - 18:30
Fri, 21 Sep 2018
10:30 - 11:30
Fri, 21 Sep 2018
10:30 - 11:30
NUS
Tomaso Aste Filtering information with networks: understanding market structure and predicting market changes Thu, 13 Sep 2018
17:00 - 18:00
Thu, 13 Sep 2018
18:00 - 19:00
Thu, 13 Sep 2018
11:00 - 12:00
Thu, 13 Sep 2018
10:00 - 11:00
UCL
Multiple Speakers Workshop on Machine Learning and Big Data Analytics Thu, 6 Sep 2018
15:30 - 18:30
Thu, 6 Sep 2018
16:30 - 19:30
Thu, 6 Sep 2018
09:30 - 12:30
Thu, 6 Sep 2018
08:30 - 11:30
HKUST
Harald Uhlig Some simple Bitcoin Economics Mon, 3 Sep 2018
16:00 - 17:30
Mon, 3 Sep 2018
17:00 - 18:30
Mon, 3 Sep 2018
10:00 - 11:30
Mon, 3 Sep 2018
09:00 - 10:30
NUS
XuDong Li Exploring the Second Order Sparsity in Large Scale Optimization Thu, 30 Aug 2018
16:00 - 17:00
Thu, 30 Sep 2018
17:00 - 18:00
Thu, 30 Sep 2018
10:00 - 11:00
Thu, 30 Sep 2018
09:00 - 10:00
NUS

 

 

Jan 2018 - Jun 2018

 Speaker   Title Date and Time  Host 
Singapore and Taiwan (UTC +8) Japan
(UTC +9)
Germany
(UTC +1/2)
Frederick Kin Hing Phoa A Study of the Influence of Articles in the Large-Scale Citation Network Thu, 3 May 2018 15:00-16:00 Thu, 3 May 2018 
16:00-17:00
 

Thu, 3 May 2018 
09:00-10:00

Sinica
Su-Yun Huang Integrating multiple random sketches for sufficient dimension reduction in large-p-small-n problems
Fri, 27 Apr 2018
15:00-16:00
Fri, 27 Apr 2018
16:00-17:00
Fri, 27 Apr 2018
09:00-10:00
Sinica
Multiple Speakers

6th NUS-USPC Workshop on Machine Learning and Fin Tech
* Partial talks will be online accessible via the SEED platform

Wed, 18 Apr 2018
Thu, 3 May 2018
Wed, 18 Apr 2018 
Thu, 3 May 2018 

Wed, 18 Apr 2018 
Thu, 3 May 2018 

NUS 

Ilija Ilievski 

Interpretable Forecasting of Financial Time Series with Deep Learning

Fri, 9 Mar 2018
13:30 - 14:00 

Fri, 9 Mar 2018
14:30 - 15:00 

Fri, 9 Mar 2018
06:30 - 07:00 

ISM

Hitoshi Iwasaki

Topic Modeling and Sentiment Analysis on Japanese Financial Analyst Reports

Fri, 9 Mar 2018
13:00 - 13:30 

Fri, 9 Mar 2018
14:00 - 14:30 

Fri, 9 Mar 2018
06:00 - 06:30 

ISM

Hao Lei

Unsupervised Probabilistic Topic Modeling

Fri, 9 Mar 2018
12:30 - 13:00 

Fri, 9 Mar 2018
13:30 - 14:00 

Fri, 9 Mar 2018
05:30 - 06:00 

ISM

 

 

 

 

 

 

Title: Current Dipole Localization from EEG with Birth-Death Process

SpeakerHideitsu Hino (Associate Professor), ISM
He received his Bachelor’s degree in engineering in 2003, and Master’s degree in Applied Mathematics and Physics in 2005 from Kyoto University.He earned Doctor’s degree in engineering in 2010 from Waseda University.He is an Associate Professor at The Institute of Statistical Mathematics.His research interest includes the analysis of learning algorithms from; the view point of geometry. He is also interested in sampling methods, kernel methods, distance metric learning, ranking models and their applications.

Time:   17:00 - 18:00 (Japan time), Wed, 9 Jan 2019

Venue:  Institute of Statistical Mathematics

Abstract:  We explore the EEG source localization problem as the estimation of current dipoles. We formulate the relation between current dipoles and EEG observation by state-space model and consider birth-death process of curret dipole. In this study, the location and moment of dipoles are estimated by Rao-Blackwellized Particle Filter (RBPF) and whether a new dipole is born or an existed dipole disappears is estimated by Bayesian Information Criterion (BIC). We propose a new dipole birth-death model for BIC model selection. By synthetic and real data experiments, we check the effectiveness of our method. This work is done in collaboration with Keita Nakamura, Noboru Murata at Waseda university, Sho Sonoda at RIKEN, Masahiro Kawasaki at university of Tsukuba, and Shotaro Akaho at AIST.

 

Event calendar 

 

 

 

Title: Integral transform methods and spectral filters for the pricing of exotic options

Speaker: Guido Germano 

Time:   09:30 - 10:30 (UK time), Fri, 14 Dec 2018

Venue:  UCL

Abstract: We present numerical methods to calculate fluctuation identities for exponential Lévy processes with discrete and continuous monitoring. This includes the Spitzer identities which give the distribution of the maximum or the minimum of a random path, the joint distribution at maturity with the extrema staying below or above a barrier, and the more difficult case of the two-barriers exit problem. These identities are given in the Fourier-z or Fourier-Laplace domain and require numerical inverse z and Laplace transforms as well as, for the required Wiener-Hopf factorisations, numerical Hilbert transforms based on a sinc function expansion and thus ultimately on the fast Fourier transform. In most cases we achieve exponential convergence with respect to the number of grid points, in some cases improving the rate of convergence with spectral filters to mitigate the Gibbs phenomenon for discrete Fourier transforms. As motivating applications we price barrier, lookback, quantile and Bermudan options.  

Event calendar 

 

 

 

Title: Workshop on Social Media Analytics

Speaker Topic
Ying Chen Topic Sentiment Asset Pricing with DNN Supervised Learning
Philip YU Detecting Comments Showing Risk for Suicide in Social Media

 

Time:   14:40 - 16:20 (HongKong time), Thu, 29 Nov 2018

Venue: HSU, Hong Kong

 

Event calendar

 

 

 

 

Title: Social closure and the evolution of cooperation via indirect reciprocity

SpeakerSimone Righi

Time:   10:00 - 11:00 (UK time), Fri, 16 Nov 2018

Venue:  UCL

Abstract:  Direct and indirect reciprocity are good candidates to explain the fundamental problem of evolution of cooperation. We explore the conditions under which different types of reciprocity gain dominance and their performances in sustaining cooperation in the PD played on simple networks. We confirm that direct reciprocity gains dominance over indirect reciprocity strategies also in larger populations, as long as it has no memory constraints. In the absence of direct reciprocity, or when its memory is flawed, different forms of indirect reciprocity strategies are able to dominate and to support cooperation. We show that indirect reciprocity relying on social capital inherent in closed triads is the best competitor among them, outperforming indirect reciprocity that uses information from any source. Results hold in a wide range of conditions with different evolutionary update rules, extent of evolutionary pressure, initial conditions, population size, and density.

 

Event calendar 

 

 

 

 

 

Title: A direct sampler from log-affine models with aid of computational algebra

Speaker: Shuhei Mano (Associate Professor), ISM
He received his Doctor's degree in mathematical physics in 1999. He has been a member of the institute of statistical mathematics from 2010. His current research interest includes Bayesian nonparametrics, algebraic methods, and inference on combinatorial stochastic processes.

Time:   17:00 - 18:00 (Japan time), Wed, 31 Oct 2018

Venue:  Institute of Statistical Mathematics

Abstract:  Multinomial sampling from log-affine models is very common in count data analyses, including the two-by-two contingency table with fixed marginal sums. The Markov chain Monte Carlo (MCMC) is a very popular methods, because it does not need normalizing constants. However, it is well known that MCMC has several drawbacks, including departure from the stationarity and auto correlation among samples. Diaconis and Sturmfels (1998, Ann. Stat.) proposed use of the theory of Groebner basis to study the Markov bases, which are bases for an MCMC sampler. Their work is one of a origin of the recent developments of algebraic statistics. In this talk, I will introduce a direct sampler from log-affaine models (M 2017, Electron. J. Stat.), which enables independent sampling from the exact distribution. It is based on the Weyl algebra with the A-hypergeometric system, which was introduced by Israel Gel'fand and coauthors in late 1980s. In this talk, I will concentrate on the two-by-two contingency table, because every statistician knows about it. I will explain details why and how we can directly sample from the distribution with aid of computational algebra. This talk is partly based on a joint work with Nobuki Takayama at Kobe university.

Event calendar 

 

 

 

 

 

 

Title: Blockchain for a Complex Society

SpeakerTomaso Aste

Time:   15:00 - 16:00 (Singapore time), Fri, 19 Oct 2018

Venue:  NUS,  6 Science Drive 2, S16-05-101

Abstract:  Blockchain is a technology that uses community validation to keep synchronized the content of ledgers replicated across multiple users. Although blockchain derives its origins from technologies introduced decades ago, it has gained popularity with Bitcoin and it is at the core of a soaring number of new cryptocurrencies and a large range of applications beyond currencies and finance. Blockchain has opened new possibilities for businesses acting as a coordination technology through which trust can be created, consensus can be reached and value can be transferred within a community without the need of intermediaries. Blockchain can radically change the way in which information is managed, value is created and rules are enforced in current socio economic systems. I’ll introduce to the fundamental concepts of blockchain technologies and guide through this new fascinating environment that is bubbling with new ideas and socio-economic applications.

Event calendar 

 



 

 

 

 

 

Title: Reciprocity and success in academic careers

SpeakerGiacomo Livan

Time:   10:00 - 11:00 (UK time), Thu, 11 Oct 2018

Venue:  UCL

Abstract:  The growing importance of citation-based bibliometric indicators in shaping the prospects of academic careers incentivizes scientists to boost the numbers of citations they receive. Whereas the exploitation of self-citations has been extensively documented, the impact of reciprocated citations has not yet been studied. In this talk I will discuss reciprocity in a citation network of academic authors, and compare it with the average reciprocity computed in a variety of null network model ensembles. I will show that obtaining citations through reciprocity correlates negatively with a successful career in the long term. Nevertheless, at the aggregate level there is evidence of a steady increase in reciprocity over the years, largely fuelled by the exchange of citations between coauthors. These results characterize the structure of author networks in a time of increasing emphasis on citation-based indicators, and I will discuss their implications towards a fairer assessment of academic impact

Event calendar 

 

 

 

  

 

 

 

 

Title: Sentimental Markets:  How Information Flow Drives Patterns in Asset Pricing

SpeakerRichard Peterson

Time:  15:00 - 16:00 (Singapore time), Thu, 11 Oct 2018

Venue:  NUS

Abstract: There is substantial and growing quantitative evidence that information flow drives asset prices across equities, currencies, commodities, and cryptocurrencies. Innovations in natural language processing, cloud computing, and the proliferation of online news and social media have yielded vast new datasets on which to study the role of information (themes and sentiments) in driving human risk taking behavior, and by proxy, asset prices.  Dr. Richard Peterson will review academic literature as well as industry best practices in examining the role of media (news, social media, and search) in driving price patterns.  Dr. Peterson will also review machine learning and statistical techniques that appear to capture significant relationships in media data. Dr. Peterson has an Electrical Engineering undergraduate degree, a medical doctorate, and completed residency training in psychiatry.  He performed postdoctoral research in Neuroeconomics at Stanford and is a board-certified psychiatrist.  As an Associate Editor of the Journal of Behavioral Finance and a data vendor (the Thomson Reuters MarketPsych Indices), Dr. Peterson presents a hybrid academic and industry understanding of asset pricing phenomena.

 

 

 

 

 

 

 

Title: Workshop on  risk analytics 

Speaker Topic
Tomohiro ANDO Quantile Co-Movement in Financial Markets
David WONG How Retail Traders Approach Risk

 

Time:   18:30 - 20:30 (HongKong time), Thu, 4 Oct 2018

Venue: HKUST 

Event calendar

 

 

 

 

Title: Flexible HAR Model for Realized Volatility

SpeakerOstap Okhrin 

Co-Authors: Francesco Audrino und Chen Huang (Uni St. Gallen)

Time:   16:30 - 17:30 (Singapore time), Fri, 28 Sep 2018

Venue:  National University of Singapore

Abstract:  The Heterogeneous Autoregressive (HAR) model is commonly used in modeling the dynamics of realized volatility. In this paper, we propose a flexible HAR(1,...,p) specification, employing the adaptive LASSO and its statistical inference theory to see whether the lag structure (1, 5, 22) implied from an economic point of view can be recovered by statistical methods. The model differs from Audrino and Knaus (2016) where the authors apply LASSO on the AR(p) model, which does not necessarily lead to a HAR model. Adaptive LASSO estimation and the subsequent hypothesis testing results fail to show strong evidence that such a fixed lag structure can be recovered by a flexible model. We also apply the group LASSO and related tests to check the validity of the classic HAR, which is rejected in most cases. The results justify our intention to use a flexible lag structure while still keeping the HAR frame. In terms of the out-of-sample forecasting, the proposed flexible specification works comparably to the benchmark HAR(1, 5, 22). Moreover, the time-varying model combinations show that when the market environment is not stable, the fixed lag structure (1, 5, 22) is not particularly accurate and effective.

Event calendar 

 

 

 

 

 

Title: Bias correction for the maximum likelihood estimator of the extreme value index

Speaker: Xuan Leng

Time:   15:00 - 16:00 (Singapore time), Fri, 14 Sep 2018

Venue: National University of Singapore

Abstract:  We conduct bias correction for the maximum likelihood estimator (MLE) of the extreme value index. Compared to the original MLE, the bias-corrected estimator allows for using a larger fraction of observations in tail region for estimation, which results in a lower asymptotic variance. The bias correction is achieved by subtracting the asymptotic bias from the original MLE, which is estimated by a two-step approach. We prove the asymptotic behavior of the proposed bias-corrected estimator. Extensive simulations show the superiority of the bias-corrected estimator compared to existing estimators of the extreme value index. We apply the bias-corrected MLE to test whether human life span is unlimited.

Event calendar 

 

 

 

 

 

Title: Filtering information with networks: understanding market structure and predicting market changes

Speaker:Tomaso Aste

Time:   10:00 - 11:00 (UK time), Thu, 13 Sep 2018

Venue:  UCL

Abstract:  We are witnessing interesting times rich of information readily available for us all. Using, understanding and filtering such information has become a major activity across science, industry and society at large.Networks are excellent tools to represent and model complex systems such as the human brain or the financial market.  Sparse networks constructed from observational data of complex systems can be used to filter information by extracting the core interaction structure in a simplified but representative way [1,2]. I will show how information filtering networks built from similarity measures, both linear and non-linear, can be used to process information while it is generated reducing complexity and dimensionality while keeping the integrity of the dataset [1-3].  I’ll describe  how predictive probabilistic models can be associate to such networks [3,4].  I will show how reliable, predictive and useful these models are to describe financial market structure and to predict regime changes [5,6]. 


Reference Publications
[1] Tumminello, Michele, Tomaso Aste, Tiziana Di Matteo, and Rosario N. Mantegna. "A tool for filtering information in complex systems." Proceedings of the National Academy of Sciences of the United States of America 102, no. 30 (2005): 10421-10426.
[2] AsteTomaso, Tiziana Di Matteo, and S. T. Hyde. "Complex networks on hyperbolic surfaces." Physica A: Statistical Mechanics and its Applications 346, no. 1-2 (2005): 20-26.
[3] Massara, Guido Previde, Tiziana Di Matteo, and Tomaso Aste. "Network filtering for big data: triangulated maximally filtered graph." Journal of complex Networks 5, no. 2 (2016): 161-178.
[4] W Barfuss, GP Massara, T Di Matteo and T Aste, "Parsimonious modeling with information filtering networks" Physical Review E, 94(6), (2016): p.062306.
[5] AsteTomaso, W. Shaw, and Tiziana Di Matteo. "Correlation structure and dynamics in volatile markets." New Journal of Physics 12, no. 8 (2010): 085009.
[6] PF Procacci & T Aste, "Forecasting market states", arXiv preprint arXiv:1807.05836 (2018).

 

Event calendar 

 

 

 

 

 

Title: Workshop on Machine Learning and Big Data Analytics

Speaker Topic
Hong YAN Co-clustering Analysis of Multidimensional Big Data
Tak CHU The Challenges of Big Data Integration
Dennis LEUNG Advancing with Artificial Intelligence

Time:   15:30 - 18:30 (HongKong time), Thu, 6 Sep 2018

Venue: HKUST

 Event calendar

 

 

 

 

 

 

Title: Some simple Bitcoin Economics

Speaker:Harald Uhlig

Time:   16:00 - 17:30 (Singapore time), Mon, 3 Sep 2018

Venue:  NUS Lim Tay Boh Seminar Room (AS2 03-12)

Abstract:  In an endowment economy, we analyze coexistence and competition between traditional fiat money (Dollar) and cryptocurrency (Bitcoin). Agents can trade consumption goods in either currency or hold on to currency for speculative purposes. A central bank ensures a Dollar inflation target, while Bitcoin mining is decentralized via proof-of-work. We analyze Bitcoin price evolution and interaction between the Bitcoin price and monetary policy which targets the Dollar. We obtain a fundamental pricing equation, which in its simplest form implies that Bitcoin prices form a martingale. We derive conditions, under which Bitcoin speculation cannot happen, and the fundamental pricing equation must hold. We explicitly construct examples for equilibria.

Event calendar 

 

 

 

 

 

Title: Exploring the Second Order Sparsity in Large Scale Optimization

Speaker: XuDong Li

Time:   16:00 - 17:00 (Singapore time), Thu, 30 Aug 2018

Venue: National University of Singapore, S16-05-98

Abstract:  In this talk, we shall demonstrate how the second order sparsity (SOS) in important optimization problems such as the sparse optimization models, semidefinite programming, and many others can be explored to induce efficient algorithms. The SOS property allows us to incorporate the semismooth Newton methods into the augmented Lagrangian method framework in a way that the subproblems involved only need low to medium costs, e.g., for lasso problems with sparse solutions, the costs for solving the subproblems at each iteration of our second order method are comparable or even lower than those in many first order methods. Consequently, with the fast convergence rate in hand, usually asymptotically superlinear linear, we now reach the stage of being able to solve many challenging large scale convex optimization problems efficiently and robustly. For the purpose of illustration, we present a highly efficient software called LassoNAL for solving the well-known Lasso-type problems.

Event calendar 

 

 

 

 

 

 

Title: A Study of the Influence of Articles in the Large-Scale Citation Network

SpeakerFrederick Kin Hing Phoa, Academia Sinica

Time:   15:00-16:00(Taiwan), Thu, 3 May 2018 

Venue:  Institute of Statisticeal Science, Academia Sinica

Abstract: Nowadays there are many research metrics at the author-, article-, journallevels, like the impact factors and many others. However, none of them possess a universally meaningful interpretation on the research influence at all levels, not mentioning that many are subject-biased and consider neighboring relations only. In this talk, we introduce a new network-based research metric called the network influence. It utilizes all information in the whole network and it is universal to any levels. Due to its statistical origin, this metric is computationally efficient and statistically interpretable even if one applies it to a large-scale network. This work demonstrates the analysis of networks via network influence using a large-scale citation database called the Web of Science. By just considering the articles among statistics community in 2005-2014, the network influence of all articles are calculated and compared, resulting in a top-ten important articles that are slightly different from the list via impact factors. This metric can be easily extended to other similar networks embedded in the Web of Science. In addition, we discuss some ongoing extension works in the Web of Science at the end. This is a joint work with Ms. Livia Lin-Hsuan Chang (ISM, Japan) and Professor Junji Nakano (ISM, Japan)

Event calendar 

 

 

 

Title: Integrating multiple random sketches for sufficient dimension reduction in large-p-small-n problems

 SpeakerSu-Yun Huang , Academia Sinica

Time:   15:00-16:00(Taiwan), Fri, 27 April 2018 

Venue:  Institute of Statisticeal Science, Academia Sinica

Abstract: Sufficient dimension reduction (SDR) is continuing an active research field nowadays. When estimating the central subspace (CS), inverse regression based SDR methods involve solving a generalized eigenvalue problem, which can be problematic under the large-p-small-n situation. In recent years, there are emerging new techniques in numerical linear algebra, called randomized algorithms or random sketching, for high dimensional and large scale problems. To overcome the large-p-small-n problem in SDR, we combine the idea of statistical inference with random sketching to propose a new SDR method, named integrated random-partition SDR (iRP-SDR). Our method consists of the following steps. (1) Randomly partition the covariates into subsets to construct an envelope subspace with low dimension. (2) Obtain a sketch estimate of the CS by applying conventional SDR method in the constructed envelope subspace. (3) Repeat the above two steps for multiple times and integrate these multiple sketches to form a final estimate of the CS. The advantageous performance of iRP-SDR is demonstrated via simulation studies and an EEG data analysis. (joint with Hung Hung, National Taiwan University)

Event calendar

 

 

 

 

 

Title: 6th NUS-USPC Workshop

Time:   Wed, 18 Apr 2018 & Thu, 19 Apr 2018 

Venue:  IMS, NUS

Abstract: 

Wed, 18 April 2018
Time Activity
09:40 – 10:25 Jean-François CHASSAGNEUX
University Paris Diderot, France
A Probabilistic Numerical Method for MFG
10:55 - 11:40 Steven KOU
National University of Singapore, Singapore
Designing Stable Coins
11:40 – 12:25 Claudio FONTANA
University Paris Diderot, France
The Value of Informational Arbitrage
14:00 – 14:45 Ilija ILIEVSKI
National University of Singapore, Singapore
Interpretable Forecasting of Financial Time Series with Deep Learning
14:45 – 15:30 Simon TRIMBORN
National University of Singapore, Singapore
Sparse-Group Network AutoRegressive Model for the Bitcoin Blockchain
16:00 – 16:45 Hao LEI
National University of Singapore, Singapore
Unsupervised Probabilistic Topic Modelling
16:45 – 17:30 Min DAI
National University of Singapore, Singapore
Robo-Advising: A Dynamic Mean-Variance Approach

 

Thu, 19 April 2018
Time Activity
09:30 – 10:30 Nicolas LANGRENE
CSIRO, Australia
Huyên PHAM
University Paris Diderot, France
Deep Learning Algorithms for Stochastic Control Problems
11:00 – 11:45 Michael KUPPER
University of Konstanz, Germany
Computation of Optimal Transport and Related Hedging Problems via Penalization and Neural Networks
11:45 – 12:30 Christa CUCHIERO
University of Vienna, Austria
Calibration Of Financial Models With Neural Networks
14:00 – 15:00 Ivan GUO
Monash University, Australia
Gregoire LOEPER
Monash University, Australia
Machine Learning in Stochastic Optimal Transport and Volatility Calibration

 

Event calendar 

 

     

 

 

 

 

 

 

 

Title: Interpretable Forecasting of Financial Time Series with Deep Learning

SpeakerIlija IlievskiNational University of Singapore

Time:  14:30-15:00, Fri, 9 March 2018

Venue:  3rd floor, ISM

Abstract: In this talk I will present our deep learning approach to forecasting financial multivariate time series which indicate the market sentiment towards a financial asset. The interpretable deep neural network reveals the essential dependence between the time series' variables, and in contrast to the widely used vector autoregressive model, the deep learning model dynamically adapts the dependence coefficients to the ever-changing market conditions. Thus, the proposed method permits the study of the inter-variable relationships which yields a better understanding of the asset's future price movements and consequently increases the profitability of the asset's trading activities. I will conclude the talk with dependence analysis and forecasting performance for financial assets from different sectors and with vastly different market capitalisations.

Event calendar

 

 

 

 

Title: Topic Modelling and Sentiment Analysis on Japanese Financial Analyst Report

Speaker: Hitoshi IwasakiNational University of Singapore

Time:  14:00-14:30, Fri, 9 March 2018

Venue:  3rd floor, ISM

Abstract: We propose an asset pricing model using natural language processing (NLP). Although a series of works has  been done on news articles and corporate disclosures, analyst reports are not studied as much as other texts due to its limited availability. We gathered 76,384 Japanese analyst reports on the stocks listed on Tokyo Stock Exchange in 2016 and 2017 and perform Latent Dirichlet Allocation (LDA) analysis to identify key topics that are influential on the stock returns on the subsequent trading days. A variety of sentiment models such as Long Short Term Memory (LSTM) and Convolutional Neural Network (CNN) are then adopted to assign sentiments based on the key topics. We conduct an empirical test with the calculated topic sentiment scores to quantify the influence of each key topic.

Event calendar

 

 

 

 

Title: Unsupervised Probabilistic Topic Modeling

SpeakerHao Lei, National University of Singapore

Time: 13:30-14:00, Fri, 9 March 2018

Venue: 3rd floor, ISM

Abstract: Probabilistic topic modeling is to extract the key information topics_ from the unstructured text data. An extensively studied and applied model is the Latent Dirichlet Allocation (LDA). One of the unsolved problems in the field is to determine the number of topics. The usual approach is to try different numbers, for example, 10, 20, 30 etc, and compare their performance on the validation dataset. In this project, we propose an automatic method to find the optimal number of topics as well as key words in each topic so that the probability distribution will be concentrated on fewer significant words. We implement to analyze financial news data comprised of 644,211 articles from 2006-04-02 to 2017-04-01.

Event calendar