Takahiro Hoshino
Takahiro Hoshino studies how to combine data affected by selection bias. His team develops statistical machine-learning methods for business and economic information. Hoshino coauthored a 2018 paper with Igari on Bayesian data combination for repeated durations with unobserved missing indicators. With Mitsuhiro, he co-authored a 2020 study of kernel canonical correlation analysis for combining datasets from multiple sources. With Shinoda, Hoshino co-authored Estimation of Local Average Treatment Effect by Data Combination at AAAI 2022.
- Now at
- Keio University and RIKEN
- Works
- 147
- Citations
- 1,378
- h-index
- 18
Publication numbers from OpenAlex, as of 14 October 2026.
Timeline
-
2018
Combining incomplete data
Hoshino coauthored a 2018 paper with Igari on Bayesian data combination for repeated durations with unobserved missing indicators.
-
2020
Kernel methods for data fusion
With Mitsuhiro, he co-authored a 2020 study of kernel canonical correlation analysis for combining datasets from multiple sources.
-
2022
Causal effects through data combination
With Shinoda, Hoshino co-authored Estimation of Local Average Treatment Effect by Data Combination at AAAI 2022.
-
As of October 2026
Economic-information fusion
His team programme combines survey data, large transaction and location datasets, and macro-level information. It studies selection bias and new data-acquisition methods for managerial and policy decisions.
Sources and credits
Photo: www.riken.jp; creator not stated. License: Not stated by source. Source: riken.jp. Cropped to a square and resized. Ask us to remove or replace this photo.