The first YUIMA Summer School on Computational and Statistical Methods for Stochastic Process & Third Yuima Workshop*

This 4 days course aims at introducing researchers, PhD students and practitioners to several aspects of modern numerical and statistical analysis of time series through the R language and, in particular, the YUIMA package.

The course covers topics of R programming, time series data handling, simulation and numerical analysis for several types of statistical models including: point processes, Hawkes processes,stochastic differential equations driven by Brownian motion with or without jumps, fractional Brownian motion and Lévy processes.

* By invitation only

Third Yuima Workshop program: see this page and schedule of talks here.

Who can benefit?

Stochastic differential equations, with or without jumps, are nowadays used as statistical models in many contexts, including but not limited to, finance, insurance, phylogenetics, genomics, political analysis, economics, migration flow analysis, social network analysis, and more. They are continuous-time models fitted on discrete sampled data. Point processes, like Compound Poisson and Hawkes processes are used in Limit Order Book (LOB) analysis in trading and finance, as well as in the analysis of rainfalls in meteorology or in earthquakes analysis in seismology. Jump and Lèvy processes extends many of the above models to a variety of statistical distributions that are able to capture stylised facts about real time series. Last, but not least, fractional processes are typical tools of climate change studies. Although the course could not cover all of the above applications, the participants can benefit from understanding the simulation and estimation techniques for these classes of stochastic process and apply to their own research field with the help of the faculties of this summer school.

Where & When?

25-28 June 2019, Brixen-Bressanone, Italy

Credits

The Third YUIMA Conference is supported by Japan Science and Technology Agency CREST JPMJCR14D7; and Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research No. 17H01702 (Scientific Research).

Summary

Course Structure

The course consists of:

  • Morning lectures: 16 x 45 minutes: Tuesday to Friday 9:00-12:30h
  • Practical computer tutorials in the afternoons: Tue, Wed, Fri 14:30-17:00h

Computer Tutorials (Lab)

You will work on the labs at your own pace in small groups with expert guidance (all lecturers from the morning sessions plus teaching assistants).

Participants are required to bring their own laptop with the most recent release versions of R and Yuima installed: we will inform all the participants about detailed installation setup before the course start. Please make sure that your computer’s hardware is sufficiently powered and that you have administrator rights to install further R packages and tools if needed.

Course Materials

The course materials for the labs and lectures will be available for download during and after the course directly from this page in the program table below..

School Programme

Tuesday, June 25

9:00-9:45 Refresh on probability & stochastic processes (Y. Koike) Lec01 Slides and R code (updated 2019-06-24)
9:45 – 10:30 Refresh on statistical inference (S.M. Iacus) Lec02 Slides and R code
10:30-11:00 Coffee break  
11:00-11:45 Basics on time series analysis (Y. Koike) Lec03 Slides and R code
11:45-12:30 Monte Carlo analysis (S.M. Iacus) Lec04 Slides and R code
12:30-14:30 Lunch break (lunch self provided by the participants)  
13:30–14:30 Installation help desk  
14:30-15:30 R programming + time series import and Date/timestamps manipulation in R (E.Guidotti) Lab01 Intro to R, Data I/O
15:30-16:00 Coffee break  
16:00-17:00 Overview of Yuima and Yuima GUI (E. Guidotti) Lab02 yuima and yuimaGUI

Wednesday, June 26

09:00–09:45 Refresh on diffusion processes (N. Yoshida) Lec05 Slides
09:45–10:30 Simulation of diffusion processes (H. Masuda) Lec06 Slides (updated 2019-06-26)
10:30–11:00 Coffee break  
11:00–11:45 QLA – Quasi-likelihood analysis (N. Yoshida) Lec07 Slides
11:45–12:30 Bayesian analysis (N. Yoshida) Lec08 Slides
12:30–14:30 Lunch break (lunch self provided by the participants)  
14:30–15:30 Simulation of diffusion processes (Y. Koike) Lab03 Slides and R code (updated 2019-06-25)
15:30–16:00 Coffee break  
16:00–17:00 Estimation for diffusion processes (Y. Koike) Lab04 Slides and R code

Thursday, June 27

 

09:00–09:45 Lévy processes: basics and simulation (H. Masuda) Lec09 Slides, R code and web page
09:45–10:30 Lévy driven SDE: basics and simulation (H. Masuda) Lec10 Slides, R code and web page
10:30–11:00 Coffee break  
11:00–11:45 Quasi-likelihood estimation of Lévy driven SDE (H. Masuda) Lec11 Slides, R code and web page
11:45–12:30 Quasi-likelihood estimation of Lévy driven SDE (H. Masuda) Lab05 Slides, R code and web page
12:30–14:30 Lunch break (lunch self provided by the participants)  

Friday, June 28

09:00–09:45 Point processes: theory and simulation (L. Mercuri) Lec12 Slides
09:45–10:30 Hawkes Processes and LOB (Limit Order Book) models (L. Mercuri) Lec13 Slides (updated 2019-06-28)
10:30–11:00 Coffee break  
11:00–11:45 Variance-covariance estimation (Y. Koike) Lec14 Slides and R code
11:45–12:30 Asymptotic expansion methods (N. Yoshida) Lec15 Slides and R code
12:30–14:30 Lunch break (lunch self provided by the participants)  
14:30–15:30 Simulation and estimation of compound Poisson and Hawkes processes (L. Mercuri) Lab06 Slides and R code
15:30–16:00 Coffee break  
16:00–17:00 Model Selection ans Lasso for SDE models (S.M. Iacus) Lec16+Lab07 Slides and R code

Lecturers and Teaching Assistants

Nakahiro Yoshida (University of Tokyo, JP), Stefano Iacus (University of Milan, IT), Hiroki Masuda (Kyushu University, JP), Lorenzo Mercuri (University of Milan, IT), Yuta Koike (University of Tokyo, JP), Emanuele Guidotti (University of Neuchâtel, CH – AlgoFinance Sagl, CH).

Third YUIMA Workshop Programme (free for students of YSS2019)

Wednesday, June 26

17:00-17:05 Modelling Translation Kinetics after mRNA Transfection Using Diffusions S. Pieschner (student flash talk)
17:05-17:30 On regularized estimation for Stochastic Differential Equations S. M. Iacus
17:30-18:00 Adaptive L^q penalized estimation for diffusion processes in Yuima package F. Iafrate

Thursday, June 27

14:30–15:00 High dimensional covariance estimation in YUIMA package Y. Koike
15:00–15:30 Molecular data integration, evolutionary trajectories and diseases P. Liò
15:30–16:00 YUIMA for simulating traits and phylogenetics in the pcmabc R package K. Bartoszek
16:00–16:30 coffee break  
16:30–17:00 Point processes modelling of limit order book events I. Muni Toke
17:00–17:30 On Lévy driven models in YUIMA H. Masuda (ask directly the speaker)
17:30–18:00 Adaptive and non-adaptive estimation of degenerate diffusion processes

N. Yoshida

(ask directly the speaker)

Friday, June 28

17:00–17:05 Asymptotic equivalence in the Le Cam sense for jump diffusions G. Gazzani (student flash talk)
17:05–17:30 COGARCH(p,q) model: simulation, estimation and an application in portfolio selection L. Mercuri
17:30–18:00 Towards coding of the asymptotic expansion formula in YUIMA E. Guidotti