Workshop: Computational Aspects of Simulation and Inference for Stochastic Processes

This two-day workshop is aimed at presenting the latest results on simulation and inference for stochastic processes and their current and prospect implementation within the YUIMA R package.

Dates:
March 27 (Tue) 9:15-12:15, 14:45-17:45
March 28 (Wed) 9:15-12:15, 14:45-17:45

Location:
Department of Economics, Business and Statistics
Via Conservatorio 7, 20122 Milan
2 Floor, Departmental Seminar Room

The attendance to the workshop is open to eveRyone interested in YUIMA and its development. A considerable part of the workshop is dedicated to discussion on further development of this Project.

This workshop is supported by:
– Department of Economics, Business and Statistics, University of Milan
– CREST Japan Science and Technology Agency
– JSPS Grant-in-Aid for Scientific Research

Hypotheses testing for SDE: theoretical and numerical results

Speaker: Alessandro De Gregorio (University of Rome)
Abstract: In this talk we consider parametric hypotheses testing for discretely observed ergodic diffusion processes. We present the different test statistics proposed in literature and recall their asymptotic properties. By a numerical analysis based on the YUIMA package, we compare the performance of the tests for small sample sizes.

Estimation in fractional Gaussian noise at high-frequency

Speaker: Alexandre Brouste (University of Le Mans)
Abstract: Asymptotic efficiency of the sequence of maximum likelihood estimators is considered in the statistical experiment implying the fractional Gaussian noise observed at high-frequency. Likelihood ratio hypothesis tests are also studied with an application to oil price modeling.

Lead-lag analysis in YUIMA package

Speaker: Yuta Koike (University of Tokyo, JST CREST)
Abstract: We present the current functions for lead-lag analysis in YUIMA package as well as some examples to illustrate their empirical applications. We also demonstrate some recent statistical methods for lead-lag analysis which we attempt to implement in future versions of YUIMA package.

yuimaGUI: A graphical user interface for computational finance based on the yuima R package

Speaker: Emanuele Guidotti (University of Milan)
Abstract: Recently, the Yuima package has been coupled with a graphical user interface, namely the YuimaGUI, which simplifies the usage of the package and allows for a complete flow of analysis: from data ingestion, to model selection and or estimation, and estimation.
The yuimaGUI package provides a user-friendly interface for YUIMA. It greatly simplifies tasks such as estimation and simulation of stochastic processes and it also includes additional tools related to quantitative finance. Some of them: data retrieval: stock prices and economic indicators, time series clustering, change point analysis and lead-lag estimation.
Moreover, yuimaGUI allows to simulate a portfolio of assets and derivatives under different scenarios and further evaluate the overall returns along with their distribution.

Statistical inference for Lévy driven SDE in YUIMA package

Speaker: Yuma Uehara (Institute of Statistical Mathematics)
Abstract: This talk is concerned with estimation for Lévy driven SDE based on high-frequency observations in YUIMA package. We give a brief review of two methods: the first one is Gaussian quasi-likelihood estimation, and the second one is iterative Jarque-Bera normality test. We also explain the implementation of their methods on YUIMA package with some numerical experiments by their current version.

Sampling stepsize in practice: fine tuning and/or estimation

Speaker: Hiroki Masuda (Kyushu University)
Abstract: We consider a class of stochastic differential equations observed at high frequency. How to assign a specific value for the stepsize parameter is of practical importance, while having not received much attention in the literature. The problem is more significant when model in question is of ergodic type where the sampling stepsize in theory has to be small but not too much. In this talk, we make a closer look at this practical problem, partly with some theoretical results.
This talk is partly based on a joint work with Shoichi Eguchi.

“yuima.law”: From mathematical representation of general Lévy processes to a numerical implementation

Speaker: Lorenzo Mercuri (University of Milan)
Abstract: We present a new class called yuima.law that refers to the mathematical description of a general Lévy process used in the formal definition of a general Stochastic Differential Equation. The final aim is to have an object, defined by the user, that contains all possible information about the Lévy process considered. This class creates a link between YUIMA and other R packages that manage specific Lévy process like for example ghyp or MixedTS available on CRAN. We present as examples the simulation and the estimation of CARMA(p,q) and Point Processes regression models.

On model selection and Lasso methods in YUIMA package

Speaker: Stefano Iacus (University of Milan)
Abstract: We present the current status and future development in YUIMA regarding model selection and Lasso methods.

New YUIMA functions to be constructed

Speaker: Nakahiro Yoshida (University of Tokyo, JST CREST)
Abstract: We discuss possible extensions of YUIMA functions, e.g., higher order asymptotic expansion; global jump filters; ratio model; penalized QLA; statistical lead-lag; correlation estimation and sparse graphical methods.