Shoki Okubo

I am an Assistant Professor of Sociology at Toyo University, Tokyo. I work on causal inference under model uncertainty — which covariates a study should adjust for, how robustness should be reported honestly, and what data-driven variable selection does to statistical inference — and on what measurement itself does to the survey data we analyse.

Three identities from current work

Where a multiverse's disagreement comes from

\[ \operatorname{Var}(\hat\tau)=\underbrace{\textstyle\sum_{g} w_g\,\operatorname{Var}(\hat\tau \mid G_g)}_{\text{within graphs: sampling + specification}}\;+\;\underbrace{\operatorname{Var}_g\,E[\hat\tau \mid G_g]}_{\text{between graphs: structural}} \]

Restrict a multiverse to the specifications each candidate causal graph licenses. The structural share \(\rho=\text{between}/\text{total}\) says how much of the spread is a disagreement about the world rather than about the model.

When an outcome predictor hurts the effect on the treated

\[ V_{\mathrm{att}}(S\cup W)-V_{\mathrm{att}}(S)=\frac{1}{p^{2}}\,E\!\left[e_S\left\{\frac{1-2e_S}{1-e_S}\,V_0(S)-2\,C_{10}(S)\right\}\right] \]

For \(W \perp\!\!\!\perp A \mid S\). Unlike the average treatment effect, adding a predictor of the untreated outcome can raise the efficiency bound of the effect on the treated whenever treatment is rare and the two arms' predictions are uncorrelated — so the optimal adjustment set is not a property of the graph alone.

Panel conditioning from a cohort's own first wave

\[ \begin{aligned} \tau^{S}_t(k) &= E[Y_t\mid S_{c:t}{=}1,\,A{=}1]\;-\;E[Y_t\mid A{=}0] \\ &\quad -\;\underbrace{\big\{E[Y_c\mid S_{c:t}{=}1,\,A{=}1]-E[Y_c\mid A{=}1]\big\}}_{\text{attrition selection, measured before any conditioning}} \end{aligned} \]

A continuing cohort's entry-wave answers are unconditioned for survivors and attriters alike, so they measure how selected the survivors are; a refreshment sample then supplies the untreated mean. The design needs no follow-up of the fresh cohort and survives non-stationary attrition.

Working papers; full statements, assumptions, and proofs are in the papers.

My methodological work sits between social-science methodology and statistics. Current projects develop causally disciplined multiverse analysis — running robustness analyses only within specifications that a stated causal model licenses, and decomposing disagreement across specifications into sampling noise and structural uncertainty — together with valid post-selection inference for outcome-adaptive covariate selection and estimand-specific theory of optimal adjustment sets. A second strand asks what repeated interviewing changes in survey answers, what can and cannot be identified about it from a panel's own design, and what an analyst should believe when it cannot be identified.

Substantively I use Japan's long-running panel surveys and administrative data to study inequality over the life course: care work and long-term care policy, including households that combine childcare with elder care; why Japan's gender gaps in pay, careers, and workplace authority have proved so immobile; what staying put does in a society where most adults never move, using residential histories linked to small-area geographic data; how employment careers and family formation are jointly produced; a programme of natural-experiment studies of Japanese social policy built on administrative data that covers every municipality; and the long-run persistence of historical institutions, studied with historical sources I digitize and release myself.

I am a member of the Japanese Life Course Panel Surveys (JLPS) team at the Institute of Social Science, The University of Tokyo, where I was on the faculty from 2018 to 2025. I hold a PhD in Human Sciences from Osaka University (2017).

ORCID 0009-0007-6925-5174 · Research interests: causal inference · social-science methodology · survey methodology and measurement error · panel data · social stratification and the life course · gender inequality · residential mobility and neighbourhood effects · social policy and natural experiments · long-run persistence and historical data construction.

Selected work

Identification assumptions and strategies for panel conditioning bias based on the potential outcomes model: a natural experiment with additional random sampling (in Japanese). Advances in Social Research, 33, 2024.
Applications of machine learning in sociology (in Japanese). Advances in Social Research, 31, 2023.
Introduction to statistical causal inference: association meets causation (in Japanese). Sociological Theory and Methods, 38(1), 2023. doi
Bai, H. & Clark, M. H., Propensity Score Methods and Applications — Japanese translation with H. Kurokawa. Kyoritsu Shuppan, 2023.

All publications →

News

2026panelcond 0.1.1, an R package implementing refreshment-sample designs for panel conditioning, released on GitHub.
2026dagmv 0.1.0, an R package for causally disciplined multiverse analysis, released on GitHub.
2026Principal investigator, JSPS KAKENHI Grant-in-Aid for Scientific Research (C) 26K05332, A large-scale analysis of dual caregiving across childcare and elder care (FY2026–2029).
2025Received the 15th Advances in Social Research Award (Japanese Association for Social Research).
2025Joined the Faculty of Sociology, Toyo University, as Assistant Professor.
2023Japanese translation of Propensity Score Methods and Applications published by Kyoritsu Shuppan.