Yoann Potiron

Associate Professor

Faculty of Business and Commerce at Keio University

YoannPotiron I am from Chambery, France. I hold a B.S. with a major in Applied Mathematics and minor in Computer Science from Universite de Lyon 1 in 2010. Moreover, I obtained a M.S. (Diplome d'Ingenieur) in Probability from Ecole Polytechnique in 2013. Finally, I graduated for my Ph.D. in Statistics from University of Chicago in 2016. I joined the Faculty of Business and Commerce at Keio University in Tokyo as a Tenure-track Assistant Professor in 2016, before obtaining my tenure in 2018 and becoming an Associate Professor in 2020. I am interested in problems arising in statistics, stochastic processes, econometrics and finance. My research is partly supported by Japanese Society for the Promotion of Science Grants-in-Aid for Early-Career Scientists "Econometric methods for high frequency data" (20K13470).

I am looking for Ph.D or postdoc students (possibly working remotedely as an exchange student). I am also looking for Master or B.S. students to work on collaborative research and/or research assistanship.





General information

Contact Information

Faculty of Business and Commerce, Keio University.
2-15-45 Mita, Minato-ku, Tokyo 108-8345
E-mail: potiron (at) fbc.keio.ac.jp
Phone: +81 (0)3 5418 6571
Office : Research building 439B (4th floor)

Fields of Interest

Statistics, Stochastic Processes, Econometrics, Finance.

Employment

Education

Ph.D., Department of Statistics, The University of Chicago.
Committee: Per Aslak Mykland, Dacheng Xiu, Dan Nicolae and Greg Lawler.
September 2013 - March 2016

Research

Published/accepted papers

  1. Estimation for high-frequency data under parametric market microstructure noise, with Simon Clinet, Annals of the Institute of Statistical Mathematics, 2021, 73, 649-669.
  2. Cointegration in high frequency data, with Simon Clinet, Electronic Journal of Statistics, 2021, 15(1), 1263-1327. Download the Python code
  3. Disentangling Sources of High Frequency Market Microstructure Noise, with Simon Clinet, Journal of Business & Economic Statistics, 2021, 39(1), 18-39. Download the supplement Download the Python codePreviously circulated under the name "A relation between the efficient, transaction and mid prices: Disentangling sources of high frequency market microstructure noise".
  4. Local Parametric Estimation in High Frequency Data, with Per Aslak Mykland, Journal of Business & Economic Statistics, 2020, 38(3), 679-692. Download the supplement Download the R code Previously circulated under the name "Estimating the Integrated Parameter of the Locally Parametric Model in High-Frequency Data".
  5. Testing if the market microstructure noise is fully explained by the informational content of some variables from the limit order book, with Simon Clinet. Journal of Econometrics, 2019, 209, 289-337. Download the Python code Previously circulated under the name "Testing if the market microstructure noise is a function of the limit order book".
  6. Efficient asymptotic variance reduction when estimating volatility in high frequency data, with Simon Clinet, Journal of Econometrics, 2018, 206, 103-142. Download the Python/R code
  7. Statistical inference for the doubly stochastic self-exciting process, with Simon Clinet, Bernoulli, 2018, 24(4B), 3469-3493. Download the supplement Download the R code
  8. Classifying patents based on their semantic content, with Antonin Bergeaud and Juste Raimbault. PLoS ONE, 2017, 12(4), e0176310. Download the R code
  9. Estimation of integrated quadratic covariation with endogenous sampling times, with Per Aslak Mykland, Journal of Econometrics, 2017, 197, 20-41. Download the R code

Papers in revision

  1. None.

Submitted papers

  1. Existence in the inverse Shiryaev problem.
  2. A tale of two time scales: applications in nonparametric Hawkes processes with Ito semimartingale baseline, with Seunghyeon Yu.
  3. Generating observation times with a hitting-boundary process in high-frequency data.

Working papers/On-going projects

  1. Estimation of latency using mutually exciting point processes, with Vladimir Volkov.
  2. Estimation of Integrated Intensity in Hawkes processes with Time-Varying Baseline, with Olivier Scaillet and Seunghyeon Yu.
  3. Latency with marks, with Vladimir Volkov.
  4. Nonparametric Granger Causality Estimation of Point Processes, with Olivier Scaillet and Seunghyeon Yu.

Communication

  1. An Hypernetwork Approach to Measure Technological Innovation, with Juste Raimbault and Antonin Bergeaud, Conference on Complex Systems, 2016, Amsterdam.

Invited/contributed talk and seminar

Referee services

Teaching

2021 Spring semester

Estimating volatility in high frequency data

The main objective of this course is to develop the skills needed to do work in the industry or research with financial data. The course aims to provide students with techniques and receipts for estimation and assessment of quality of financial models. Each student is expected to choose a project, and to make a report and a final presentation to the class. In addition, the student is expected to discuss about the advancement of the project at least once during the semester. The presentation can include theory, numerical simulations and/or data analysis.

Essentials of Regression Analysis Using R

The emphasis of this class is on linear models with R. The objective is to learn what methods are available and more importantly, when they should be applied. Many examples are presented to clarify the use of the techniques and to demonstrate what conclusions can be made.

Older courses

Stochastic Calculus, An Introduction with Application

The course starts with a quick introduction to conditional expectation. Then, normal distribution, multivariate normal distribution and Brownian motion are defined and discussed carefully. Exercises and homework are based on a setup of high-frequency financial data estimation problems. This includes providing some basic tools of asymptotic statistics on the way.

専攻演習S

The main objective of this course is to develop the skills needed to conduct work in the industry or empirical research in fields operating with time series data using the software R. The course aims to provide students with techniques and receipts for estimation and assessment of quality of economic models with time series data. Each student is expected to choose a dataset, to implement some methods from the book studied in class and to make a report based on the analysis of the results. More information about the project will be provided later in the class.

Time series analysis

The main objective of this course is to develop the skills needed to do work in the industry or empirical research in fields operating with time series data. The course aims to provide students with techniques and receipts for estimation and assessment of quality of economic models with time series data. Special attention will be placed on limitations and pitfalls of different methods and their potential fixes. The course will also emphasize recent developments in Time Series Analysis and will present some open questions and areas of ongoing research. We will be using the software R, but students can do their homework using their own software.

Students

  1. Julian Kota Kikuchi (B.S., graduated 08/2022, now pre-doctoral program at the Federal Reserve Bank of Richmond)
  2. Renyi Qu (B.S., graduated 08/2022, now Master in Data Science at University of Pennsylvania)
  3. Seunghyeon Yu (Ph.D. candidate in finance at the Korea Advanced Institute of Science and Technology, expected to graduate in 02/2023)
  4. Taro Tsuchiya (B.S., graduated 08/2021, now Ph.D. student in C.S. at Carnegie Mellon University)
  5. Meihuazi Chen (B.S., graduated 08/2020, now Data Scientist at Bloomberg in Hong Kong)
  6. Kentaro Asaba (B.S., graduated 08/2020, now Trader at Societe Generale in Tokyo)