About

Jessie Q Li is an Associate Professor of Economics at the University of California, Santa Cruz. His research focuses on econometrics, with particular emphasis on constrained extremum estimators, bootstrap methods for finite-dimensional regularized estimators, and inference techniques for treatment effects under limited overlap. His work includes developing methods such as the Proximal Bootstrap, Rate-Adaptive Bootstrap for GMM, and exploring asymptotics of cross-validation, among other advanced econometric topics. Professor Li has contributed to the theoretical and applied aspects of econometrics, including estimation using Laplace transformation, penalization, MCMC, and the numerical bootstrap. His research often involves addressing issues related to misspecification, finite population treatment effects, and the development of inference procedures for complex models. His publications appear in leading journals such as the Journal of Econometrics, Econometric Theory, and The Annals of Statistics, reflecting his significant contributions to the field of econometrics.

Research topics

• Computer Science
• Statistics
• Mathematics
• Applied mathematics
• Artificial Intelligence
• Mathematical analysis
• Econometrics
• Mathematical optimization

Selected publications

• The proximal bootstrap for constrained estimators

  Journal of Statistical Planning and Inference · 2024-10-30

  article1st authorCorresponding
  PublisherDOI

• A - 3 The Relationship between Memory and Depression In an Elderly Chinese Cohort: a Baseline Analysis

  Archives of Clinical Neuropsychology · 2024-09-12

  article1st authorCorresponding

  Abstract Objective Explore the relationship between depression and memory aged 60 years and older while controlling for age, gender, and education. Method This study utilized data from the Characterization of Dementia in Asian-Pacific Islanders study by Bruce Miller, M.D., and Howie Rosen, M.D., at the Memory and Aging Center at UCSF. Inclusion criteria included participants age 60+ who completed the Benson Figure Test, GDS-30, and the Chinese Version Verbal Learning Test (CVVLT) in English, Cantonese, or Mandarin. Exclusion criteria included participants under age 60 and or experience history of neurological strokes or injuries, and disabilities related to memory. The final sample yielded 136 participants, including 41 (30.1%) males, 95 (69.9%) females, 120 (88.2%) right-handed, 3 (2.2%) left-handed, and 7 (5.1%) ambidextrous. Average age of participants was 71.46 (SD = 7.26). A memory composite score was computed based on average of performances on the Benson Figure Task delay and the CVVLT 10-minute delay. Raw GDS-30 scores represented depression. Results There was a weak positive correlation between memory composite scores, gender, and education, and a weak negative correlation between memory composite scores, age, and GDS-30 scores. Conclusions Findings indicate some evidence for a relationship between depression and memory in elderly Chinese populations. However, the sample was skewed, and consisted of less depressed and higher memory functioning participants. Despite these limitations, it is pertinent to explore how healthcare providers may outreach to this population to monitor risk factors, improve timeliness of dementia diagnosis, and maximize treatment of reversible causes, adherence to treatment plans, and help families care plan.

  PublisherDOI

• Adjuvant therapy and recurrence risk in non-myoinvasive high-grade (stage IC) endometrial cancer: A systematic review and meta-analysis

  Gynecologic Oncology · 2024-09-20 · 4 citations

  review
  PublisherDOI

• RATE-ADAPTIVE BOOTSTRAP FOR POSSIBLY MISSPECIFIED GMM

  Econometric Theory · 2024-01-22 · 3 citations

  articleSenior authorCorresponding

  We consider inference for possibly misspecified GMM models based on possibly nonsmooth moment conditions. While it is well known that misspecified GMM estimators with smooth moments remain $\sqrt {n}$ consistent and asymptotically normal, globally misspecified nonsmooth GMM estimators are $n^{1/3}$ consistent when either the weighting matrix is fixed or when the weighting matrix is estimated at the $n^{1/3}$ rate or faster. Because the estimator’s nonstandard asymptotic distribution cannot be consistently estimated using the standard bootstrap, we propose an alternative rate-adaptive bootstrap procedure that consistently estimates the asymptotic distribution regardless of whether the GMM estimator is smooth or nonsmooth, correctly or incorrectly specified. Monte Carlo simulations for the smooth and nonsmooth cases confirm that our rate-adaptive bootstrap confidence intervals exhibit empirical coverage close to the nominal level.

  PublisherDOI

• Prognostic Effect of Mismatch Repair Status in Early-Stage Endometrial Cancer Treated With Adjuvant Radiation: A Multi-institutional Analysis

  International Journal of Radiation Oncology*Biology*Physics · 2024-01-21 · 3 citations

  articleOpen access

  PURPOSE: The aim of this work was to report the effect of mismatch repair (MMR) status on outcomes of patients with stage I-II endometrioid endometrial adenocarcinoma (EEC) who receive adjuvant radiation therapy. METHODS AND MATERIALS: This is a multi-institutional retrospective cohort study across 11 institutions in North America. Patients with known MMR status and stage I-II EEC status postsurgical staging were included. Overall survival (OS) and recurrence-free survival (RFS) rates were estimated via the Kaplan-Meier method. Univariable and multivariable analyses were performed via Cox proportional hazard models for RFS and OS. Statistical analyses were conducted using SPSS version 27. RESULTS: In total, 744 patients with a median age at diagnosis of 65 years (IQR, 58-71) were included. Most patients were White (69.4%) and had Federation of Obstetrics and Gynecology 2009 stage I (84%) and Federation of Obstetrics and Gynecology grade 1 to 2 (73%). MMR deficiency was reported in 234 patients (31.5%), whereas 510 patients (68.5%) had preserved MMR. External beam radiation therapy with or without vaginal brachytherapy was delivered to 186 patients (25%), whereas 558 patients (75%) received vaginal brachytherapy alone. At a median follow-up of 43.5 months, the estimated crude OS and RFS rates for the entire cohort were 92.5% and 84%, respectively. MMR status was significantly correlated with RFS. RFS was inferior for MMR deficiency compared with preserved MMR (74.3% vs 88.6%, P < .001). However, no difference in OS was seen (90.8% vs 93.2%, P = .5). On multivariable analysis, MMR deficiency status was associated with worse RFS (hazard ratio, 1.86; P = .001) but not OS. CONCLUSIONS: MMR status was independently associated with RFS but not OS in patients with early-stage EEC who were treated with adjuvant radiation therapy. These findings suggest that differential approaches to surveillance and/or treatment based on MMR status could be warranted.

  PublisherOA PDFDOI

• Weeding Out the Culprit: Cannabinoid-Associated Stevens-Johnson Syndrome

  Cureus · 2023-05-24 · 1 citations

  articleOpen access1st authorCorresponding

  We describe a case of Stevens-Johnson syndrome (SJS) in a 32-year-old female who initially presented with a several-day history of worsening rash. Diagnosis of cannabinoid-associated SJS was established following skin biopsy and detailed history-taking of medication and other recreational drug usages. The patient was treated with pain management, antihistamines, and topical steroids with no complications following discharge. There currently exists limited literature describing SJS due to recreational drug usage.

  PublisherOA PDFDOI

• Asymptotics of K-Fold Cross Validation

  Journal of Artificial Intelligence Research · 2023-11-14 · 12 citations

  articleOpen access1st authorCorresponding

  This paper investigates the asymptotic distribution of the K-fold cross validation error in an i.i.d. setting. As the number of observations n goes to infinity while keeping the number of folds K fixed, the K-fold cross validation error is √ n-consistent for the expected out-of-sample error and has an asymptotically normal distribution. A consistent estimate of the asymptotic variance is derived and used to construct asymptotically valid confidence intervals for the expected out-of-sample error. A hypothesis test is developed for comparing two estimators’ expected out-of-sample errors and a subsampling procedure is used to obtain critical values. Monte Carlo simulations demonstrate the asymptotic validity of our confidence intervals for the expected out-of-sample error and investigate the size and power properties of our test. In our empirical application, we use our estimator selection test to compare the out-of-sample predictive performance of OLS, Neural Networks, and Random Forests for predicting the sale price of a domain name in a GoDaddy expiry auction.

  PublisherOA PDFDOI

• Capnocytophaga aortitis: A dog’s gift to its owner

  IDCases · 2023-01-01 · 1 citations

  articleOpen access

  We present a case of Capnocytophaga aortitis in an 82-year-old male with fever, weakness, confusion, and back pain. Diagnosis was established following a ruptured abdominal aortic aneurysm and subsequent blood culture growth of Capnocytophaga species. He was treated with endovascular aortic repair in addition to a six-week course of ceftriaxone followed by long-term antibiotic suppression with amoxicillin-clavulanate. Capnocytophaga aortitis is exceedingly rare and poorly described in current literature.

  PublisherDOI

• PO17

  Brachytherapy · 2022-11-01

  article
  PublisherDOI

• GPP07  Presentation Time: 9:40 AM

  Brachytherapy · 2022-11-01

  article1st authorCorresponding
  PublisherDOI

Frequent coauthors

• Han Hong

  36 shared

• Michael P. Leung

  University of California, Santa Cruz

  27 shared

• A. Ronald Gallant

  Pennsylvania State University

  25 shared

• Yoonjung Kang

  Ministry of Food and Drug Safety

  4 shared

• Shari Damast

  4 shared

• Karina Kung

  University of Toronto

  4 shared

• Jasmine Yeung

  The Scarborough Hospital

  4 shared

• Michael A. Dyer

  4 shared