Associate Professor
Theoretical Statistics and Probability Theory
Research interests
Statistics for stochastic processes, and its applications to high-frequency financial data
Current research

I am studying statistical inferences for continuous-time stochastic processes observed at a high-frequency in a fixed interval. Since such a model typically appears as high-frequency financial data, it is actively studied in financial econometrics. In the meantime, from a mathematical point of view, its study requires stochastic calculus and limit theorems for semimartingales as central tools, so there are also many researchers on statistical inferences for stochastic processes in this area. I am one of such researchers. More concretely, I am studying estimation of the covariance structure of a continuous-time stochastic process from its high-frequency observation data. In such a model, the problems of non-synchronous observations, microstructure noise and jumps have been studied for the past two decades as major issues. However, the central part of these problems has been resolved in recent studies, so the researchers turn to refining the details. Although I am interested in such a problem, I am recently studying how to model and estimate lead-lag relationships between two continuous-time stochastic processes. I am also interested in the relationship between high-frequency data analysis and high-dimensional statistics.

Selected publications
  1. Koike, Y.: An estimator of the cumulative co-volatility of asynchronously observed semimartingales with jumps, Scandinavian Journal of Statistics 41, 460-481 (2014).
  2. Koike, Y.: Limit theorems for the pre-averaged Hayashi-Yoshida estimator with random sampling, Stochastic Processes and their Applications 124, 2699-2753 (2014).
  3. Koike, Y: Estimation of integrated covariances in the simultaneous presence of nonsynchronicity, microstructure noise and jumps, Econometric Theory 32, 533-611 (2016).
  4. Koike, Y: Quadratic covariation estimation of an irregularly observed semimartingale with jumps and noise, Bernoulli 22, 1894-1936 (2016).
  5. Koike, Y.: Time endogeneity and an optimal weight function in pre-averaging covariance estimation, Statistical Inference for Stochastic Processes 20, 15-56 (2017).

Memberships, activities and


Japan Statistical Society

Research Associate Professor at Tokyo Metropolitan University