arXiv subjects

Akira Okazaki

Publications and source records attributed to Akira Okazaki.

9 recordsLinked to original sources

arXiv2026

Groupwise Predictor Envelope Models for Multivariate Linear Regression

Envelope methods improve estimation efficiency in multivariate analysis by isolating low-dimensional structures that contain all the information material to the parameter of interest. In multivariate linear regression with random predictors, predictor envelope models achieve this goal by removing variation in the predictors that is immaterial to the regression. In many applications, observations are naturally divided into several groups, such as treatment groups, regions, or demographic strata, and the regression relationship may differ between groups. Motivated by this setting, we propose a groupwise predictor envelope model for multivariate linear regression. The proposed model assumes that the group-specific regression coefficient matrices are represented through a common predictor envelope subspace while allowing group-specific regression effects and group-specific error covariance matrices. We derive an objective function for estimating the common predictor envelope, obtain the corresponding regression estimators, and establish asymptotic normality together with an explicit asymptotic variance formula. Moreover, we show that the proposed estimator is asymptotically more efficient than the estimator obtained by fitting predictor envelope models separately to each group. This theoretical advantage over the existing work is also demonstrated through simulation studies.

stat.ME↗

arXiv2026

Mixed-effects Outcome-Adaptive Lasso for Propensity Score Estimation under Partial Interference

Interference occurs when one individual's treatment or exposure affects another individual's outcome. In particular, we assume partial interference, where individuals are divided into groups such that there is no interference between individuals in different groups. In observational studies, inverse probability weighting (IPW) based on propensity scores is often used for causal effect estimation. However, under partial interference, the group-level propensity score must be estimated, and it is more likely to take extreme values than the usual individual-level propensity score. As a result, IPW estimators may have large variances. This problem can become more serious when many covariates are available. In this study, we propose an Outcome-Adaptive Lasso based on a mixed-effects logistic regression model to stably estimate causal effects under partial interference. The proposed method performs covariate selection and estimation in the propensity score model simultaneously while accounting for unobserved group-level heterogeneity in treatment assignment. Under regularity conditions, we show that the proposed method has the oracle property and that the IPW estimators based on the proposed method are consistent and asymptotically normal. Through Monte Carlo simulations, we demonstrate that the proposed method tends to select confounders and prognostic factors at high frequencies, while excluding instrumental variables and spurious variables. The results further suggest that the proposed method improves the finite-sample efficiency of IPW estimators. We evaluate the performance of the proposed method using malaria data from the Democratic Republic of the Congo Demographic and Health Survey (DHS).

stat.ME↗

arXiv2026

Variable Fusion and Selection via a Spike-and-Slab Approach with Nonlocal Priors

Variable fusion in linear regression models is a statistical method that identifies covariates making similar contributions to the response variable and imposes the same coefficient values on them. Many methods for variable fusion also incorporate variable selection for practical reasons. In this paper, within the Bayesian model averaging (BMA) framework, we propose a spike-and-slab-based Bayesian method that performs both variable fusion and selection. This is challenging in the BMA framework because one must construct a discrete model space that accommodates both selection and fusion and assign suitable priors over that space. In the proposed method, we present a way to explore a model space for variable fusion and selection based on Gibbs sampling by devising a prior distribution for latent variables representing the model. Furthermore, among non-local priors with superior model selection properties, we construct a prior tailored for variable fusion and use it as the slab distribution. We examine the effectiveness of the proposed method through theoretical and empirical studies.

stat.ME↗

arXiv2024

Multi-task learning via robust regularized clustering with non-convex group penalties

Multi-task learning (MTL) aims to improve estimation and prediction performance by sharing common information among related tasks. One natural assumption in MTL is that tasks are classified into clusters based on their characteristics. However, existing MTL methods based on this assumption often ignore outlier tasks that have large task-specific components or no relation to other tasks. To address this issue, we propose a novel MTL method called Multi-Task Learning via Robust Regularized Clustering (MTLRRC). MTLRRC incorporates robust regularization terms inspired by robust convex clustering, which is further extended to handle non-convex and group-sparse penalties. The extension allows MTLRRC to simultaneously perform robust task clustering and outlier task detection. The connection between the extended robust clustering and the multivariate M-estimator is also established. This provides an interpretation of the robustness of MTLRRC against outlier tasks. An efficient algorithm based on a modified alternating direction method of multipliers is developed for the estimation of the parameters. The effectiveness of MTLRRC is demonstrated through simulation studies and application to real data.

stat.ME↗

arXiv2023

Multi-Task Learning Regression via Convex Clustering

Multi-task learning (MTL) is a methodology that aims to improve the general performance of estimation and prediction by sharing common information among related tasks. In the MTL, there are several assumptions for the relationships and methods to incorporate them. One of the natural assumptions in the practical situation is that tasks are classified into some clusters with their characteristics. For this assumption, the group fused regularization approach performs clustering of the tasks by shrinking the difference among tasks. This enables us to transfer common information within the same cluster. However, this approach also transfers the information between different clusters, which worsens the estimation and prediction. To overcome this problem, we propose an MTL method with a centroid parameter representing a cluster center of the task. Because this model separates parameters into the parameters for regression and the parameters for clustering, we can improve estimation and prediction accuracy for regression coefficient vectors. We show the effectiveness of the proposed method through Monte Carlo simulations and applications to real data.

stat.ME↗

arXiv2021

Multi-task Learning for Compositional Data via Sparse Network Lasso

A network lasso enables us to construct a model for each sample, which is known as multi-task learning. Existing methods for multi-task learning cannot be applied to compositional data due to their intrinsic properties. In this paper, we propose a multi-task learning method for compositional data using a sparse network lasso. We focus on a symmetric form of the log-contrast model, which is a regression model with compositional covariates. The effectiveness of the proposed method is shown through simulation studies and application to gut microbiome data.

stat.ME↗

arXiv2013

Phase Variation of Earthshine Polarization Spectra

We present the results of the optical spectropolarimetry of Earthshine on the Moon for Earth phase angles ranging from 49 to 96 degrees. The observations were conducted on 2011 March 9-13 (UT) using the spectropolarimeter HBS installed on the 1.88 m telescope at Okayama Astrophysical Observatory. The wavelength coverage was 450-850 nm with a resolution of 6 nm. The observed Earthshine polarization degree spectra exhibit decreasing polarization degree with increasing wavelength at any phase. The overall degree of polarization increases as the Earth approaches a quadrature phase. The phase dependence differs with the wavelengths; the maximum polarization for the V band occurs at a phase angle of ~90 degrees, whereas that for longer wavelengths is reached at larger phase angles. This is interpreted as indicating that Earthshine polarization at shorter wavelengths is dominated by atmospheric Rayleigh scattering, whereas that at longer wavelengths has an increasingly effective contribution from the Earth surface reflection. The wavelength dependence of the phase angle of the maximum polarization appears to be unique among the terrestrial planetary bodies in the Solar System. Therefore this might constitute important evidence pointing toward a distinctive characteristic of the Earth: the planet has a scattering but transparent atmosphere above its surface.

astro-ph.EP↗

arXiv2006

Spectropolarimetric Study on Circumstellar Structure of Microquasar LS I +61deg 303

We present optical linear spectropolarimetry of the microquasar LS I +61$^{\circ}$ 303. The continuum emission is mildly polarized (up to 1.3 %) and shows almost no temporal change. We find a distinct change of polarization across the H$α$ emission line, indicating the existence of polarization component intrinsic to the microquasar. We estimate the interstellar polarization (ISP) component from polarization of the H$α$ line and derive the intrinsic polarization component. The wavelength dependence of the intrinsic component is well explained by Thomson scattering in equatorial disk of the Be-type mass donor. The position angle (PA) of the intrinsic polarization $\sim 25^{\circ}$ represents the rotational axis of the Be disk. This PA is nearly perpendicular to the PA of the radio jet found during quiescent phases. Assuming an orthogonal disk-jet geometry around the compact star, the rotational axis of the accretion disk is almost perpendicular to that of the Be disk. Moreover, according to the orbital parameters of the microquasar, the compact star is likely to get across the Be disk around their periastron passage. We discuss the peculiar circumstellar structure of this microquasar inferred from our observation and possible connection with its high-energy activities.

astro-ph↗

arXiv2006

Low- and Medium-Dispersion Spectropolarimetry of Nova V475 Sct (Nova Scuti 2003): Discovery of an Asymmetric High-Velocity Wind in a Moderately Fast Nova

We present low-resolution ($R\sim 90$) and medium-resolution ($R\sim 2500$) spectropolarimetry of Nova V475 Sct with the HBS instrument, mounted on the 0.91-m telescope at the Okayama Astrophysical Observatory, and with FOCAS, mounted on the 8.2-m Subaru telescope. We estimated the interstellar polarization toward the nova from the steady continuum polarization components and H$α$ line emission components. After subtracting the interstellar polarization component from the observations, we found that the H$α$ emission seen on 2003 October 7 was clearly polarized. In the polarized flux spectrum, the H$α$ emission had a distinct red wing extending to $\sim +4900$ km s$^{-1}$ and a shoulder around $+3500$ km s$^{-1}$, showing a constant position angle of linear polarization $θ_{\rm *}\simeq 155\arcdeg\pm 15\arcdeg$. This suggests that the nova had an asymmetric outflow with a velocity of $v_{\rm wind}\simeq 3500$ km s$^{-1}$ or more, which is six times higher than the expansion velocity of the ionized shell at the same epoch. Such a high-velocity component has not previously been reported for a nova in the `moderately fast' speed class. Our observations suggest the occurrence of violent mass-loss activity in the nova binary system even during the common-envelope phase. The position angle of the polarization in the H$α$ wing is in good agreement with that of the continuum polarization found on 2003 September 26 ($p_{\rm *}\simeq 0.4$--0.6 %), which disappeared within the following 2 d. The uniformity of the PA between the continuum polarization and the wing polarization on October 7 suggests that the axis of the circumstellar asymmetry remained nearly constant during the period of our observations.

astro-ph↗