Searcharxiv⌕ Search

arXiv subjects

Nakahiro Yoshida

Publications and source records attributed to Nakahiro Yoshida.

At least 19 recordsLinked to original sources
arXiv2026

Risk comparison theorems and application to deep learning of diffusion coefficients

We investigate the nonparametric estimation of the diffusion matrix in stochastic differential equations featuring multidimensional, strong mixing covariate processes. We propose a flexible statistical framework based on general function classes that does not require a linear basis representation, rendering our results directly applicable to deep neural network estimators. Our approach employs a two-step estimation procedure: constructing a preliminary nonparametric quasi-likelihood estimator and subsequently regularizing it via a $β$-Hölder class approximation. We establish general risk comparison theorems between empirical and generalization risks for arbitrary estimators without relying on a specific probabilistic structure of the underlying process. In diffusion matrix learning based on $n + 1$ observations over the time interval $[0,T]$, the derived upper bounds capture the intrinsic interplay between the $T$-rate, associated with the mixing behavior of the covariate process, and the intrinsic $n$-rate governing the volatility estimation.

math.ST↗
arXiv2026

Drift estimation for rough processes under small noise asymptotic : trajectory fitting method

We consider a process $X^\ve$ that solves a stochastic Volterra equation with an unknown parameter $θ^\star$ in the drift function. The Volterra kernel is singular, and includes as an example, $K\_0(u)=c u^{α-1/2} \id{u>0}$ with $α\in (0,1/2)$. It is assumed that the diffusion coefficient is proportional to $\ve \to 0$. From an observation of the path $(X^\ve\_s)\_{s\in[0,T]}$, we construct a Trajectory Fitting Estimator, which is shown to be consistent and asymptotically normal. We also specify identifiability conditions insuring the $L^p$ convergence of the estimator.

math.ST↗
arXiv2026

Drift estimation for rough processes under small noise asymptotic : QMLE approach

We consider a process $X^\ve$ solution of a stochastic Volterra equation with an unknown parameter $θ^\star$ in the drift function. The Volterra kernel is singular near zero, exhibiting a behavior comparable to $K\_0(u)=cu^{α-1} \id{u>0}$ with $α\in (1/2,1)$.It is assumed that the diffusion coefficient is proportional to $\ve \to 0$. Based on discrete observations, with a mesh size $h\to0$, of the Volterra process, we construct a Quasi Maximum Likelihood Estimator. The main step is to assess the error arising in the reconstruction of the path of a semimartingale from the inversion of the Volterra kernel. We show that this error decreases as $h^{1/2}$ regardless of the value of $α$. Then, we can introduce an explicit contrast function, which yields an efficient estimator when $\ve \to 0$.

math.ST↗
arXiv2026

Statistical inference for highly correlated stationary point processes and noisy bivariate Neyman-Scott processes

Motivated by estimating the lead-lag relationships in high-frequency financial data, we propose noisy bivariate Neyman-Scott point processes with gamma kernels (NBNSP-G). NBNSP-G tolerates noises that are not necessarily Poissonian and has an intuitive interpretation. Our experiments suggest that NBNSP-G can explain the correlation of orders of two stocks well. A composite-type quasi-likelihood is employed to estimate the parameters of the model. However, when one tries to prove consistency and asymptotic normality, NBNSP-G breaks the boundedness assumption on the moment density functions commonly assumed in the literature. Therefore, under more relaxed conditions, we show consistency and asymptotic normality for bivariate point process models, which include NBNSP-G. Our numerical simulations also show that the estimator is indeed likely to converge.

math.ST↗
arXiv2025

Deep learning of point processes for modeling high-frequency data

We investigate applications of deep neural networks to a point process having an intensity with mixing covariates processes as input. Our generic model includes Cox-type models and marked point processes as well as multivariate point processes. An oracle inequality and a rate of convergence are derived for the prediction error. A simulation study shows that the marked point process can be superior to the simple multivariate model in prediction. We apply the marked ratio model to real limit order book data

math.ST↗
arXiv2024

Log-rank test with coarsened exact matching

It is of special importance in the clinical trial to compare survival times between the treatment group and the control group. Propensity score methods with a logistic regression model are often used to reduce the effects of confounders. However, the modeling of complex structures between the covariates, the treatment assignment and the survival time is difficult. In this paper, we consider coarsened exact matching (CEM), which does not need any parametric models, and we propose the weighted log-rank statistic based on CEM. We derive asymptotic properties of the weighted log-rank statistic, such as the weak convergence to a Gaussian process in Skorokhod space, in particular the asymptotic normality, under the null hypothesis and the consistency of the log-rank test. Simulation experiments are also conducted to compare the performance of the log-rank statistic with a propensity score method and CEM. Simulation studies show that the log-rank statistic based on CEM is more robust than the log-rank statistic based on the propensity score.

math.ST↗
arXiv2024

Asymptotic expansion of the drift estimator for the fractional Ornstein-Uhlenbeck process

We present an asymptotic expansion formula of an estimator for the drift coefficient of the fractional Ornstein-Uhlenbeck process. As the machinery, we apply the general expansion scheme for Wiener functionals recently developed by the authors [26]. The central limit theorem in the principal part of the expansion has the classical scaling T^{1/2}. However, the asymptotic expansion formula is a complex in that the order of the correction term becomes the classical T^{-1/2} for H in (1/2,5/8), but T^{4H-3} for H in [5/8, 3/4).

math.PR↗
arXiv2023

Asymptotic expansion for batched bandits

In bandit algorithms, the randomly time-varying adaptive experimental design makes it difficult to apply traditional limit theorems to off-policy evaluation of the treatment effect. Moreover, the normal approximation by the central limit theorem becomes unsatisfactory for lack of information due to the small sample size of the inferior arm. To resolve this issue, we introduce a backwards asymptotic expansion method and prove the validity of this scheme based on the partial mixing, that was originally introduced for the expansion of the distribution of a functional of a jump-diffusion process in a random environment. The theory is generalized in this paper to incorporate the backward propagation of random functions in the bandit algorithm. Besides the analytical validation, the simulation studies also support the new method. Our formulation is general and applicable to nonlinearly parametrized differentiable statistical models having an adaptive design.

stat.ME↗
arXiv2023

Penalized estimation for non-identifiable models

We derive asymptotic properties of penalized estimators for singular models for which identifiability may break and the true parameter values can lie on the boundary of the parameter space. Selection consistency of the estimators is also validated. The problem that the true values lie on the boundary is dealt with by our previous results that are applicable to singular models, besides, penalized estimation and non-ergodic statistics. In order to overcome non-identifiability, we consider a suitable penalty such as the non-convex Bridge and the adaptive Lasso that stabilizes the asymptotic behavior of the estimator and shrinks inactive parameters. Then the estimator converges to one of the most parsimonious values among all the true values. In particular, the oracle property can also be obtained even if parametric structure of the singular model is so complex that likelihood ratio tests for model selection are labor intensive to perform. Among many potential applications, the examples handled in the paper are: (i) the superposition of parametric proportional hazard models and (ii) a counting process having intensity with multicollinear covariates.

math.ST↗
arXiv2022

Quasi-maximum likelihood estimation and penalized estimation under non-standard conditions

The purpose of this article is to develop a general parametric estimation theory that allows the derivation of the limit distribution of estimators in non-regular models where the true parameter value may lie on the boundary of the parameter space or where even identifiability fails. For that, we propose a more general local approximation of the parameter space (at the true value) than previous studies. This estimation theory is comprehensive in that it can handle penalized estimation as well as quasi-maximum likelihood estimation under such non-regular models. Besides, our results can apply to the so-called non-ergodic statistics, where the Fisher information is random in the limit, including the regular experiment that is locally asymptotically mixed normal. In penalized estimation, depending on the boundary constraint, even the Bridge estimator with $q<1$ does not necessarily give selection consistency. Therefore, some sufficient condition for selection consistency is described, precisely evaluating the balance between the boundary constraint and the form of the penalty. Examples handled in the paper are: (i) ML estimation of the generalized inverse Gaussian distribution, (ii) quasi-ML estimation of the diffusion parameter in a non-ergodic Itô process whose parameter space consists of positive semi-definite symmetric matrices, while the drift parameter is treated as nuisance and (iii) penalized ML estimation of variance components of random effects in linear mixed models.

math.ST↗
arXiv2022

Asymptotic expansion of an estimator for the Hurst coefficient

Asymptotic expansion is presented for an estimator of the Hurst coefficient of a fractional Brownian motion. For this, a recently developed theory of asymptotic expansion of the distribution of Wiener functionals is applied. The effects of the asymptotic expansion are demonstrated by numerical studies.

math.ST↗
arXiv2022

Order estimate of functionals related to fractional Brownian motion and asymptotic expansion of the quadratic variation of fractional stochastic differential equation

We derive an asymptotic expansion for the quadratic variation of a stochastic process satisfying a stochastic differential equation driven by a fractional Brownian motion, based on the theory of asymptotic expansion of Skorohod integrals converging to a mixed normal limit. In order to apply the general theory, it is necessary to estimate functionals that are a randomly weighted sum of products of multiple integrals of the fractional Brownian motion, in expanding the quadratic variation and identifying the limit random symbols. To overcome the difficulty, we introduce two types of exponents by means of the "weighted graphs" capturing the structure of the sum in the functional, and investigate how the exponents change by the action of the Malliavin derivative and its projection.

math.PR↗
arXiv2021

Simplified quasi-likelihood analysis for a locally asymptotically quadratic random field

The asymptotic decision theory by Le Cam and Hajek has been given a lucid perspective by the Ibragimov-Hasminskii theory on convergence of the likelihood random field. Their scheme has been applied to stochastic processes by Kutoyants, and today this plot is called the IHK program. This scheme ensures that asymptotic properties of an estimator follow directly from the convergence of the random field if a large deviation estimate exists. The quasi-likelihood analysis (QLA) proved a polynomial type large deviation (PLD) inequality to go through a bottleneck of the program. A conclusion of the QLA is that if the quasi-likelihood random field is asymptotically quadratic and if a key index reflecting identifiability the random field has is non-degenerate, then the PLD inequality is always valid, and as a result, the IHK program can run. Many studies already took advantage of the QLA theory. However, not a few of them are using it in an inefficient way yet. The aim of this paper is to provide a reformed and simplified version of the QLA and to improve accessibility to the theory. As an example of the effects of the theory based on the PLD, the user can obtain asymptotic properties of the quasi-Bayesian estimator by only verifying non-degeneracy of the key index.

math.ST↗
arXiv2021

Global jump filters and realized volatility

For a semimartingale with jumps, we propose a new estimation method for integrated volatility, i.e., the quadratic variation of the continuous martingale part, based on the global jump filter proposed by Inatsugu and Yoshida [8]. To decide whether each increment of the process has jumps, the global jump filter adopts the upper $α$-quantile of the absolute increments as the threshold. This jump filter is called global since it uses all the observations to classify one increment. We give a rate of convergence and prove asymptotic mixed normality of the global realized volatility and its variant "Winsorized global volatility". By simulation studies, we show that our estimators outperform previous realized volatility estimators that use a few adjacent increments to mitigate the effects of jumps.

math.ST↗
arXiv2021

Global jump filters and quasi-likelihood analysis for volatility

We propose a new estimation scheme for estimation of the volatility parameters of a semimartingale with jumps based on a jump-detection filter. Our filter uses all of data to analyze the relative size of increments and to discriminate jumps more precisely. We construct quasi-maximum likelihood estimators and quasi-Bayesian estimators, and show limit theorems for them including $L^p$-estimates of the error and asymptotic mixed normality based on the framework of the quasi-likelihood analysis. The global jump filters do not need a restrictive condition for the distribution of the small jumps. By numerical simulation we show that our "global" method obtains better estimates of the volatility parameter than the previous "local" methods.

stat.ME↗
arXiv2021

Divergence of an integral of a process with small ball estimate

The paper contains sufficient conditions on the function $f$ and the stochastic process $X$ that supply the rate of divergence of the integral functional $\int_0^Tf(X_t)^2dt$ at the rate $T^{1-ε}$ as $T\to\infty$ for every $ε>0$. These conditions include so called small ball estimates which are discussed in detail. Statistical applications are provided.

math.PR↗
arXiv2020

Asymptotic expansion of a variation with anticipative weights

Asymptotic expansion of a variation with anticipative weights is derived by the theory of asymptotic expansion for Skorohod integrals having a mixed normal limit. The expansion formula is expressed with the quasi-torsion, quasi-tangent and other random symbols. To specify these random symbols, it is necessary to classify the level of the effect of each term appearing in the stochastic expansion of the variable in question. To solve this problem, we consider a class ${\cal L}$ of certain sequences $({\cal I}_n)_{n\in{\mathbb N}}$ of Wiener functionals and we give a systematic way of estimation of the order of $({\cal I}_n)_{n\in{\mathbb N}}$. Based on this method, we introduce a notion of exponent of the sequence $({\cal I}_n)_{n\in{\mathbb N}}$, and investigate the stability and contraction effect of the operators $D_{u_n}$ and $D$ on ${\cal L}$, where $u_n$ is the integrand of a Skorohod integral. After constructed these machineries, we derive asymptotic expansion of the variation having anticipative weights. An application to robust volatility estimation is mentioned.

math.PR↗