Kei Nakagawa
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Kei Nakagawa has a total of 75 co-authors. Here are the top co-authors by number of shared publications:
- Ryuta Sakemoto
Faculty of Economics and Business Hokkaido University Sapporo Hokkaido Japan
8 shared publications Mitsuyoshi Imamura
6 shared publications- Yusuke Uchiyama
MAZIN Inc., 3-29-14 Nishi-Asakusa, Tito city, Tokyo 111-0035, Japan
5 shared publications - Masaya Abe
Innovation Lab Nomura Asset Management Co.,Ltd Japan
4 shared publications Kenichi Yoshida
University of Tsukuba, Graduate School of Business Sciences
3 shared publicationsMasahiro Kato
3 shared publications
Latest Publications
- 26 October 2026
- Published by Elsevier in Finance Research Letters
- Vol. 114
International equity factor performance: Do risk and mispricing labels add information?
- 26 October 2026
- Published by Elsevier in Finance Research Letters
- Vol. 114
International equity factor performance: Do risk and mispricing labels add information?
- Open Access
- 16 October 2026
- Published by Elsevier in SSRN Electronic Journal
Weak Factors for Commodity Futures Portfolios
This study investigates whether uncertainty prices the cross-section of commodity futures portfolios. We employ commodity futures portfolios sorted by ten different characteristics and estimate risk premia on uncertainty indices using supervised principal component analysis, which is more robust to the presence of weak factors. We find that the estimated risk premia on non-text-based uncertainty variables are statistically significant and have theoretically consistent signs over the full sample period, in a joint estimation with seven latent factors that includes both tradable and nontradable factors. Some of these risk factors are also linked to individual commodity futures. In contrast, we do not find significant risk premia on text-based factors, such as policy uncertainty and geopolitical risk, using either cross-sectional commodity futures portfolios or individual commodity futures.
- 16 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessWeak Factors for Commodity Futures Portfolios
This study investigates whether uncertainty prices the cross-section of commodity futures portfolios. We employ commodity futures portfolios sorted by ten different characteristics and estimate risk premia on uncertainty indices using supervised principal component analysis, which is more robust to the presence of weak factors. We find that the estimated risk premia on non-text-based uncertainty variables are statistically significant and have theoretically consistent signs over the full sample period, in a joint estimation with seven latent factors that includes both tradable and nontradable factors. Some of these risk factors are also linked to individual commodity futures. In contrast, we do not find significant risk premia on text-based factors, such as policy uncertainty and geopolitical risk, using either cross-sectional commodity futures portfolios or individual commodity futures.
- 26 August 2026
- Published by Springer Nature in AI & SOCIETY
- p. 1-17
Governing the AI-mediated corporation: corporate governance and the limits of detached AI governance
- 26 August 2026
- Published by Springer Nature in AI & SOCIETY
- p. 1-17
Governing the AI-mediated corporation: corporate governance and the limits of detached AI governance
- 5 August 2026
- Published by Elsevier in Pacific-Basin Finance Journal
- Vol. 100
Index fund quality is persistent, but investors underreact: Evidence from Japan
- 5 August 2026
- Published by Elsevier in Pacific-Basin Finance Journal
- Vol. 100
Index fund quality is persistent, but investors underreact: Evidence from Japan
- Open Access
- 29 July 2026
- Published by Elsevier in SSRN Electronic Journal
Subspace Regularized Principal Component Analysis Using Prior Exposure Information
Rolling PCA is useful for multi-asset factor analysis but often produces unstable factor spaces that rotate across estimation windows. This paper proposes subspace-regularized PCA, which shrinks the estimated factor space toward an economically motivated prior exposure subspace while preserving rotational invariance. In multi-asset data, moderate regularization reduces mediumhorizon subspace rotation and factor-portfolio turnover with limited loss of reconstruction fit. High-dimensional tests using Fama-French size and book-tomarket sorted portfolios show similar stability-fit-turnover trade-offs. A longonly minimum-variance exercise further shows that the stabilized factor model preserves risk performance while modestly reducing portfolio turnover and drawdown.
- 29 July 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessSubspace Regularized Principal Component Analysis Using Prior Exposure Information
Rolling PCA is useful for multi-asset factor analysis but often produces unstable factor spaces that rotate across estimation windows. This paper proposes subspace-regularized PCA, which shrinks the estimated factor space toward an economically motivated prior exposure subspace while preserving rotational invariance. In multi-asset data, moderate regularization reduces mediumhorizon subspace rotation and factor-portfolio turnover with limited loss of reconstruction fit. High-dimensional tests using Fama-French size and book-tomarket sorted portfolios show similar stability-fit-turnover trade-offs. A longonly minimum-variance exercise further shows that the stabilized factor model preserves risk performance while modestly reducing portfolio turnover and drawdown.
- 1 July 2026
- Published by Elsevier in Finance Research Letters
- Vol. 108
Subspace regularized principal component analysis using prior exposure information
- 1 July 2026
- Published by Elsevier in Finance Research Letters
- Vol. 108
Subspace regularized principal component analysis using prior exposure information
- Open Access
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 1227-1231
Risk-Aware Utility Re-Ranking for Financial Asset Recommendation
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 1227-1231
Open AccessRisk-Aware Utility Re-Ranking for Financial Asset Recommendation
- Open Access
Nansde-net: A neural sde framework for generating time series with memory
Modeling time series with long- or short-memory characteristics is a fundamental challenge in many scientific and engineering domains. While fractional Brownian motion has been widely used as a noise source to capture such memory effects, its incompatibility with Itô calculus limits its applicability in neural stochastic differential equation~(SDE) frameworks. In this paper, we propose a novel class of noise, termed Neural Network-kernel ARMA-type noise~(NA-noise), which is an Itô-process-based alternative capable of capturing both long- and short-memory behaviors. The kernel function defining the noise structure is parameterized via neural networks and decomposed into a product form to preserve the Markov property. Based on this noise process, we develop NANSDE-Net, a generative model that extends Neural SDEs by incorporating NA-noise. We prove the theoretical existence and uniqueness of the solution under mild conditions and derive an efficient backpropagation scheme for training. Empirical results on both synthetic and real-world datasets demonstrate that NANSDE-Net matches or outperforms existing models, including fractional SDE-Net, in reproducing long- and short-memory features of the data, while maintaining computational tractability within the Itô calculus framework.
Open AccessNansde-net: A neural sde framework for generating time series with memory
Modeling time series with long- or short-memory characteristics is a fundamental challenge in many scientific and engineering domains. While fractional Brownian motion has been widely used as a noise source to capture such memory effects, its incompatibility with Itô calculus limits its applicability in neural stochastic differential equation~(SDE) frameworks. In this paper, we propose a novel class of noise, termed Neural Network-kernel ARMA-type noise~(NA-noise), which is an Itô-process-based alternative capable of capturing both long- and short-memory behaviors. The kernel function defining the noise structure is parameterized via neural networks and decomposed into a product form to preserve the Markov property. Based on this noise process, we develop NANSDE-Net, a generative model that extends Neural SDEs by incorporating NA-noise. We prove the theoretical existence and uniqueness of the solution under mild conditions and derive an efficient backpropagation scheme for training. Empirical results on both synthetic and real-world datasets demonstrate that NANSDE-Net matches or outperforms existing models, including fractional SDE-Net, in reproducing long- and short-memory features of the data, while maintaining computational tractability within the Itô calculus framework.