Hau-Hung Yang

• Ph.D. in PsychologySep. 2022 – Jun. 2026

  National Taiwan University, Taipei, Taiwan (R.O.C.)

  • GPA: 4.19 / 4.3
  • Advisor: Prof. Yung-Fong Hsu
• M.S. in PsychologySep. 2020 – Jun. 2022

  National Taiwan University, Taipei, Taiwan (R.O.C.)

  • GPA: 4.13 / 4.3
• B.S. in PsychologySep. 2016 – Jun. 2020

  National Taiwan University, Taipei, Taiwan (R.O.C.)

  • GPA: 4.12 / 4.3

Precision and Robustness in Adaptive Testing: An Item Response Theory Analysis within an Online Convex Optimization Framework

Advisor: Prof. Yung-Fong Hsu · Completed: June 2026

Computerized adaptive testing (CAT) assigns items adaptively to estimate ability with as few items as possible. Despite a rich asymptotic literature, a framework for understanding finite-sample behavior has been missing. This dissertation brings the Online Convex Optimization (OCO) framework to CAT, organized around two complementary themes.

Precision

Establishes sublinear information regret for the classical information-maximization method under stated model and regularity conditions, and constructs anytime-valid confidence intervals via martingale concentration inequalities and confidence sequences.

Robustness

Develops confidence-set UCB, which optimizes item information over a confidence set for examinee ability. Simulations evaluate sensitivity to item-parameter error; joint estimation of ability and item parameters during online calibration is a future extension.

• NTU Presidential Award for Graduate Students (臺大研究生校長獎)2026

  National Taiwan University

  Awarded for outstanding academic and research achievement.

• Professor Chao-ming Cheng Memorial Scholarship2026

  Department of Psychology, National Taiwan University

  Departmental graduate-student award for outstanding research, academic year 2025–2026.

• Dean's Award2026

  College of Science, National Taiwan University

  Awarded for outstanding research achievement, academic year 2025–2026.

• Covered by The New York Times, Dec. 18, 2025 — “Birthrates Are Falling, but Don’t Blame Dogs in Strollers” (Amanda Taub); also Marginal Revolution, Dec. 2025

  Coverage of: Chen, K.-M., Lin, M.-J., Yang, H.-H., & Yen, S. Y.-H. (2025). Cats, Dogs, and Babies: Quasi-Experimental Evidence on Substitutes or Complements from Linked Administrative Records. SSRN working paper.

• Postdoctoral ScholarAug 2026 – present

  Institute of Statistical Science, Academia Sinica, Taipei, Taiwan (R.O.C.)

  Government Statistics Research Group. Policy research with Ministry of the Interior administrative data; research question and empirical design in development.

• Full-time Research Assistant (R&D Substitute Military Service)July 2023 – February 2025

  Department of Economics, National Taiwan University

• On the Estimation of Standard Errors for the Ability Estimate in Item Response Models (NSTC 113-2410-H-002-217)Aug. 2024 – Jul. 2025

  PI: Prof. Yung-Fong Hsu

  Concerns the estimation of standard errors for the ability estimate in IRT models, advancing estimation methods for computerized adaptive testing. My work covers algorithm development, simulation experiments, literature review, and manuscript preparation; the dissertation work that followed is listed under Dissertation.

• Family, Health, and Markets I–III (NSTC 2628 series)Aug. 2022 – Jul. 2025

  PI: Prof. Kuan-Ming Chen

  Investigates interactions among individuals, families, and the market under health shocks and long-term care demands. I analyze government administrative data to study wage inequality in Taiwan with the Abowd–Kramarz–Margolis (AKM) framework, and intra-household decision-making and tax-filing strategies using comprehensive income-tax records, complemented by surveys and field experiments.

• Gender, Fertility, Children, and Health I (NSTC 114-2628-H-002-019)Aug. 2025 – Jul. 2026

  PI: Prof. Kuan-Ming Chen

  Quasi-experimental analysis of pets and fertility using linked pet-registration and birth records with lottery windfalls as an income shock (working paper “Cats, Dogs, and Babies”; covered by The New York Times, December 2025). I handled record linkage and estimation.

• Evaluation of an Education Program Using Item Response TheoryJun. 2022 – Dec. 2025

  PI: Prof. Yu-Chang Chen

  Evaluates the treatment effect of PaGamO, an online learning platform, using a 2-PL IRT model. I review the IRT literature, conduct algorithmic research, and implement the algorithms; we are developing a Bayesian state-space framework to estimate user ability and account for non-linear improvement over time.

• Issue-Oriented Big Data Analyses on Social Inclusion and Smart Governance I–II (NSTC 2425 series)Jan. 2023 – Dec. 2024

  PI: Prof. Jia-Chung Fan (PI); Prof. Kuan-Ming Chen (co-PI)

  Examines the substitutability/complementarity between infants and pets, using fertility decisions as a behavioral lens. The study links administrative data from Taiwan's Ministry of Finance (tax records) and Ministry of Agriculture (pet registration data).

• Neural Correlates of Emotion as an Arbitrator to Reconcile the Conflicts within Dual Processing in the Context of Decision Making Under RiskAug. 2020 – Jul. 2022

  PI: Prof. Yung-Fong Hsu

  Behavioral and fMRI experiments probing the role of emotion when people face dual-process induced cognitive conflict under risk. I analyzed fMRI data and contributed to developing an estimation method for utility-function parameters.

• Analysis of Cognitive Structures Underlying Financial BehaviorJun. 2020 – Nov. 2021

  PI: Prof. Tsung-Ren Huang

  Sought common factors explaining financial behavioral biases. I reviewed the bias-related literature and analyzed behavioral data using exploratory factor analysis.

• From Mind Reading to Mind Sharing: Neural Correlates of Cognitive and Affective Theory of Mind and Their Applications to Salesforce EnhancementSep. 2018 – Jun. 2020

  PI: Prof. Heng-Chiang Huang

  Reviewed and reported literature on synchronization, structural equation modeling, and theory of mind, and mentored new team members in preprocessing and statistical analysis of behavioral and neuroimaging data.

• Coalition without Trust: Intra-Brain Connectivity and Inter-Brain Synchronization of Herd Behaviors in an Economic Bubble GameSep. 2017 – Jun. 2020

  PI: Prof. Yu-Ping Chen

  Used fMRI hyperscanning to explore theory-of-mind processes when participants competed in a bubble stock-market game. I reviewed literature spanning the diffusion model, reinforcement learning, prospect theory, and behavioral economics, and analyzed the fMRI and behavioral data.

Journal Articles

1. Yang, H.-H., & Hsu, Y.-F. (2024). The generalized Robbins–Monro process and its application to psychophysical experiments for threshold estimation. Journal of Mathematical Psychology, 120--121, 102855. [doi]
2. Yang, H.-H., & Hsu, Y.-F. (2024). A note on the application of stochastic approximation to computerized adaptive testing. Behaviormetrika, 51(1), 259-276. [doi]
3. Cheng, C., Yang, H.-H., & Hsu, Y.-F. (2025). Identifiability of polychoric models with latent elliptical distributions. Psychometrika, 90(2), 757–778. [doi]
4. Chen, K.-M., Chen, Y.-C., Hung, C.-C., & Yang, H.-H. (2026). Institution or major? Understanding student preferences in college admissions. Economic Inquiry. [doi]
5. Chen, K.-M., Tsai, L.-T., & Yang, H.-H. (2026). Labor Market Sorting in Taiwan. Taiwan Economic Review, 54(2), 267–289. [doi]
6. Yang, H.-H., Ho, T.-Y., & Hsu, Y.-F. (2026). Enhancing Random Forest performance on imbalanced data using the General Condorcet Model. Journal of Classification (Accepted, forthcoming). [link]

Submitted Papers

1. Yang, H.-H., Wei, C.-M., & Chen, Y.-C. (2025). On the consistency of Bayesian adaptive testing under the three-parameter logistic model. Under revision; earlier arXiv title: On the consistency of Bayesian adaptive testing under the Rasch model. [link]
2. Chang, W.-C., Yang, H.-H., Lin, T.-Y., Tong, J. C. H., & Hsu, Y.-F. (2025). Bisection-Based Elicitation of Indifference Points: Simulation Comparison and the Halved Method (HaB) as a Monotonicity Diagnostic. Under revision; Wei-Chen Chang and Hau-Hung Yang contributed equally (co-first authors).

Preprints and Manuscripts

1. Yang, H.-H., Hsu, Y.-F., & Cheng, C. (2025). Parametric bootstrapping of standard errors for the ability parameter in IRT models: Asymptotic and small sample analyses. PsyArXiv preprint. [doi]
2. Chen, K.-M., Lin, M.-J., Yang, H.-H., & Yen, S. Y.-H. (2025). Cats, Dogs, and Babies: Quasi-Experimental Evidence on Substitutes or Complements from Linked Administrative Records. SSRN working paper. [link]
   Covered by The New York Times, Dec. 18, 2025 — “Birthrates Are Falling, but Don’t Blame Dogs in Strollers” (Amanda Taub); also Marginal Revolution, Dec. 2025
3. Cheng, C., Yang, H.-H., & Hsu, Y.-F. (2024). On the existence and uniqueness of MLE of the latent trait under log-concave distributional assumptions in IRT. PsyArXiv preprint. [doi]
4. Yang, H.-H., & Hsu, Y.-F. (2024). Applying a heterogeneous random utility model to data from conjoint experiments. PsyArXiv preprint. [doi]
5. Tong, J. C. H., Yang, H.-H., & Chen, Y.-P. (2022). Decision-making process of retail investors during a financial bubble: An fMRI study. Working paper.

1. Yang, H.-H. (2026). Invited departmental seminar (title to be announced). Department of Statistics and Data Science, National Cheng Kung University, Tainan, Taiwan. Invited talk, scheduled, October 22, 2026.

2026

1. Yang, H.-H., Hsu, Y.-F., & Cheng, C. (2026). Finite-Sample Evaluation of Computerized Adaptive Testing. Poster, International Meeting of the Psychometric Society (IMPS) 2026, Seoul, South Korea. [link]
2. Cheng, C., Yang, H.-H., Hsu, Y.-F., & Lay, K.-L. (2026). Validity Evidence for the Joint Thurstonian Models. Oral presentation, International Meeting of the Psychometric Society (IMPS) 2026, Seoul, South Korea. [link]

2025

1. Cheng, C., Yang, H.-H., Hsu, Y.-F., & Lay, K.-L. (2025). Paradigmatic Decisions for Measuring Choices and Preferences: Likert Scales, Comparative Judgments, or Combined Models?: Using Academic Attribution as an Example. Poster, The 89th Annual Convention of the Japanese Psychological Association, Sendai, Japan. [link]
2. Cheng, C., Yang, H.-H., & Hsu, Y.-F. (2025). When the test information curve misleads: Exact bias, SE, and RMSE curves in item response theory. Oral presentation, The 53rd Annual Meeting of the Behaviormetric Society, Kanagawa, Japan. [link]
3. Yang, H.-H., Wei, C.-M., & Chen, Y.-C. (2025). On the consistency of Bayesian adaptive testing under the Rasch model. Oral presentation, 8th International Conference on Econometrics and Statistics (EcoSta) 2025, Tokyo, Japan. [link]
4. Cheng, C., Lay, K.-L., Yang, H.-H., & Hsu, Y.-F. (2025). A methodology for addressing the biases resulted in ipsative placement of Q-sort items. Poster, SRCD 2025 Biennial Meeting, Minneapolis, MN, USA. [link]
5. Cheng, C., Yang, H.-H., & Hsu, Y.-F. (2025). The universal Cramér–Rao lower bound: A genuine efficiency criterion for both biased and unbiased estimators. Oral presentation, 64th Annual Convention of the Taiwanese Psychological Association, Tainan, Taiwan.

2024

1. Cheng, C., Yang, H.-H., Hsu, Y.-F., & Lay, K.-L. (2024). Two are better than one: Incorporating Likert scale and comparative judgment data in joint Thurstonian models. Oral presentation, 63rd Annual Meeting of the Taiwanese Psychological Association, Taipei, Taiwan. [link]
2. Cheng, C., Yang, H.-H., & Hsu, Y.-F. (2024). On the existence and uniqueness of polychoric correlations. Poster, 88th Annual Convention of the Japanese Psychological Association, Kumamoto, Japan. [link]
3. Cheng, C., Yang, H.-H., & Hsu, Y.-F. (2024). On the existence, uniqueness, and finiteness of MLE of the ability in IRT. Poster, International Meeting of the Psychometric Society (IMPS) 2024, Prague, Czech Republic. [link]
4. Cheng, C., Yang, H.-H., Lay, K.-L., & Hsu, Y.-F. (2024). Predetermined R²: An index of later development determined by the past. Poster, 27th Biennial Meeting of the International Society for the Study of Behavioural Development (ISSBD), Lisbon, Portugal. [link]
5. Chen, K.-M., Chen, Y.-C., Hung, C.-C., & Yang, H.-H. (2024). Estimating preferences for college programs using Taiwanese college-admission data. Oral presentation, 18th International Symposium on Econometric Theory and Applications (SETA), Taipei, Taiwan. [link]
6. Chen, K.-M., Chen, Y.-C., Hung, C.-C., & Yang, H.-H. (2024). Estimating preferences for college programs using Taiwanese college-admission data. Oral presentation, Annual Conference of the Population Association of Taiwan, Taipei, Taiwan.

2023

1. Chen, K.-M., Chen, Y.-C., Hung, C.-C., & Yang, H.-H. (2023). College and major preference estimation using Taiwanese college-admission data. Oral presentation, Asian and Australasian Society of Labour Economics Conference (AASLE) 2023, Taipei, Taiwan. [link]
2. Hsu, Y.-F., & Yang, H.-H. (2023). On the consistency and relative efficiency of a generalized Robbins–Monro process for threshold estimation. Poster, Annual Meeting of the Society for Mathematical Psychology , Amsterdam, Netherlands. [link]
3. Yang, H.-H., Cheng, C., & Hsu, Y.-F. (2023). Estimates of standard errors and confidence intervals for the ability parameter in the 2PL model with fast and exact parametric bootstrap. Poster, International Meeting of the Psychometric Society (IMPS) 2023, College Park, MD, USA. [link]
4. Cheng, C., Yang, H.-H., & Hsu, Y.-F. (2023). Identifiability of polychoric models with latent elliptical distributions. Oral presentation, International Meeting of the Psychometric Society (IMPS) 2023, College Park, MD, USA. [link]

2021

1. Yang, H.-H., & Hsu, Y.-F. (2021). Modification of the estimation method of threshold: A new adaptive method taking response time and response confidence into account. Oral presentation, 2021 Annual Meeting of the Taiwanese Psychological Association, Taipei, Taiwan.

• Grant for Graduate Students to Attend International Academic ConferencesJuly 2026

  National Science and Technology Council, Taiwan (NSTC 115-2922-I-002-376) — International Meeting of the Psychometric Society (IMPS) 2026, Seoul, South Korea

• NSTC Doctoral Research ScholarshipAugust 2025 – July 2026

  National Science and Technology Council, Taiwan — Awarded through National Taiwan University

• Grant for Graduate Students to Attend International Academic ConferencesJuly 2023

  National Science and Technology Council, Taiwan (NSTC 112-2922-I-002-355) — International Meeting of the Psychometric Society (IMPS) 2023, College Park, MD, USA

• 資訊揭露成本對於不確定情境下選擇行為的影響July 2019 – February 2020

  Ministry of Science and Technology, Taiwan (MOST. 108-2813-C-002-115-H) — College Student Research Scholarship

• Workshop on AI Applications in Teaching (Session 2)Sep. 21, 2026

  Lead instructor · Taiwan Economic Association, co-hosted by the Departments of Public Finance and Economics, National Chengchi University · co-taught with Prof. Yu-Chang Chen

  Three-hour session for university instructors in economics and related fields (capped at 40).

• Workshop on AI Applications in Teaching (Session 1)Aug. 24, 2026

  Lead instructor · Taiwan Economic Association, co-hosted by the College of Social Sciences, National Taiwan University · co-taught with Prof. Yu-Chang Chen

  Full-day workshop for university instructors in economics and related fields (capped at 40): language-model fundamentals, a course-specific AI teaching assistant, and specification-driven development of teaching materials.

In total, approximately 70 participants across the two sessions: faculty from universities across Taiwan and postdoctoral researchers from Academia Sinica.

• PSY7001 Experimental Design — Fall 2024
• PSY5033 Applied Linear Regression — Spring 2023
• PSY7001 Experimental Design — Fall 2022
• PSY3010 Intermediate Statistics — Fall 2021
• PSY1004 Statistics in Psychology and Education II — Spring 2021
• PSY1003 Statistics in Psychology and Education I — Fall 2020
• PSY1004 Statistics in Psychology and Education II — Spring 2020
• PSY1003 Statistics in Psychology and Education I — Fall 2019

Analytical

• Statistical modeling — PCA, factor analysis, generalized linear models
• Psychometrics — structural equation modeling, item response theory
• Mathematical psychology — adaptive methods (stochastic approximation), drift diffusion models
• Neuroimaging analysis (fMRI)

Applied research

• Large-scale administrative data — Ministry of Finance income-tax records, Ministry of Education college-admission records, Council of Agriculture pet-registration records, Ministry of the Interior administrative data (current)
• Record linkage and reproducible analysis pipelines

Programming

• R
• Python
• MATLAB
• SAS
• LaTeX
• Stata

Software

• clt-interactive-course — interactive Central Limit Theorem teaching tool
• Generalized_RM_supplementary — R code for the generalized Robbins–Monro paper
• RF-GCM-Public — random forest aggregation with the General Condorcet Model
• eduspec-sdd-template — specification-driven course-material template