6 research outputs found
Learning structure in nested logit models
No full textThis electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.Thesis: S.M. in Transportation, Massachusetts Institute of Technology, Department of Civil and Environmental Engineering, 2019Thesis: S.M., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2019Cataloged from student-submitted PDF version of thesis.Includes bibliographical references (pages 67-68).This work is about developing an estimation procedure for nested logit models that optimizes over the nesting structure in addition to the model parameters. Current estimation practices require an a priori specification of a nesting structure. We formulate the problem of learning an optimal nesting structure as a mixed integer nonlinear programming (MINLP) optimization problem and solve it using a variant of the linear outer approximation algorithm. We demonstrate that it is indeed possible to recover the nesting structure directly from the data by applying our method to synthetic and real datasets.by Youssef Medhat Aboutaleb.S.M. in TransportationS.M.S.M.inTransportation Massachusetts Institute of Technology, Department of Civil and Environmental EngineeringS.M. Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Scienc
Theory-constrained Data-driven Model Selection, Specification, and Estimation: Applications in Discrete Choice Models
No full textThis thesis provides a framework, along with demonstrated applications, for carefully bringing data-driven flexibility to the specification and model selection of discrete choice models; while, at the same time, maintaining usability for analysis. Assumptions brought to bear under the classical theory-based paradigm enjoy varying degrees of credibility. Some are rooted in economic theory (e.g., utility maximizing behavior) or in information available to the scientist on the data generating process (e.g., exogeneity). These assumptions can be argued to be highly credible. Others are driven by convenience, convention, pursuit of smaller standard errors, or an otherwise lack of systematic specification and model selection process (e.g., restrictive functional and distributional forms, and trial-and-error specification testing). These assumptions are arguably less credible.
Our goal is to overcome some of the arbitrary specification and model selection practices that undermine credibility. To this end, theory-constrained data-driven flexibility in specification is introduced to discrete choice models through an optimization framework. Systematic data-driven methods for model selection are used to enhance replicability. The introduced flexibility is constrained to guarantee trustworthiness of predictions through consistency with theory. At the same time, the imposed constraints are validated through hypothesis tests to maintain credibility.
The framework we introduce well positions us to realize synergies between the data-driven and theory-based paradigms. The starting point for our approach is discrete choice models with well-established theoretical underpinnings that facilitate causal and behavioral interpretations. Discrete choice models consistent with random utility maximization, for example, are tethered to microeconomics and enable sound economic and welfare valuations. Further, the entire machinery of econometrics remains applicable to address endogeneity issues. This is in contrast to emerging trends in the literature that start with data-driven classifiers in pursuit of predictive gains, and then, as an afterthought, attempt to reconcile output with theory.
We provide applications of our proposed framework in addressing specification aspects of both the systematic and stochastic components of discrete choice models. Specialized solution algorithms are developed for each application– leveraging some of the latest advances in mixed-integer and conic optimization (for classical estimation) and in Markov Chain Monte Carlo methods (for Bayesian inference). The methods developed are tested for consistency using synthetic data and applied to empirical data.Ph.D
A novel global urban typology framework for sustainable mobility futures
No full textUrban mobility signficantly contributes to global carbon dioxide emissions. Given the rapid expansion and growth in urban areas, cities thus require innovative policies to ensure efficient and sustainable mobility. Urban typologies can serve as a vehicle for understanding dynamics of cities, which exhibit high variability in form, economic output, mobility behavior, among others. Yet, typologies relevant for sustainable urban mobility analyses are few, outdated and not large enough in scope. In this paper, we present a new typologization spanning 331 cities in 124 countries. Our sample represents 40\% of the global urban population and contains the most recent data from 2010 to date. Using a factor analytic and agglomerative clustering approach, we identify 9 urban factors and 12 typologies. We discuss the implications of this new framework for researchers and planners and investigate the relationships between mobility and environmental sustainability indicators. Notably, we show an immediate application of the urban typologies to better understanding travel behavior and also describe their usage for detailed large-scale simulation in representative prototype cities for insights into sustainable future mobility policy pathways. Our data and results are publicly available for further exploration and will serve as a foundation for future analyses toward desirable urban and environmental outcomes
Effects of hospital facilities on patient outcomes after cancer surgery: an international, prospective, observational study
No full text© 2022 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY 4.0 licenseBackground: Early death after cancer surgery is higher in low-income and middle-income countries (LMICs) compared with in high-income countries, yet the impact of facility characteristics on early postoperative outcomes is unknown. The aim of this study was to examine the association between hospital infrastructure, resource availability, and processes on early outcomes after cancer surgery worldwide. Methods: A multimethods analysis was performed as part of the GlobalSurg 3 study—a multicentre, international, prospective cohort study of patients who had surgery for breast, colorectal, or gastric cancer. The primary outcomes were 30-day mortality and 30-day major complication rates. Potentially beneficial hospital facilities were identified by variable selection to select those associated with 30-day mortality. Adjusted outcomes were determined using generalised estimating equations to account for patient characteristics and country-income group, with population stratification by hospital. Findings: Between April 1, 2018, and April 23, 2019, facility-level data were collected for 9685 patients across 238 hospitals in 66 countries (91 hospitals in 20 high-income countries; 57 hospitals in 19 upper-middle-income countries; and 90 hospitals in 27 low-income to lower-middle-income countries). The availability of five hospital facilities was inversely associated with mortality: ultrasound, CT scanner, critical care unit, opioid analgesia, and oncologist. After adjustment for case-mix and country income group, hospitals with three or fewer of these facilities (62 hospitals, 1294 patients) had higher mortality compared with those with four or five (adjusted odds ratio [OR] 3·85 [95% CI 2·58–5·75]; p<0·0001), with excess mortality predominantly explained by a limited capacity to rescue following the development of major complications (63·0% vs 82·7%; OR 0·35 [0·23–0·53]; p<0·0001). Across LMICs, improvements in hospital facilities would prevent one to three deaths for every 100 patients undergoing surgery for cancer. Interpretation: Hospitals with higher levels of infrastructure and resources have better outcomes after cancer surgery, independent of country income. Without urgent strengthening of hospital infrastructure and resources, the reductions in cancer-associated mortality associated with improved access will not be realised. Funding: National Institute for Health and Care Research
Global variation in postoperative mortality and complications after cancer surgery: a multicentre, prospective cohort study in 82 countries
No full text© 2021 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY-NC-ND 4.0 licenseBackground: 80% of individuals with cancer will require a surgical procedure, yet little comparative data exist on early outcomes in low-income and middle-income countries (LMICs). We compared postoperative outcomes in breast, colorectal, and gastric cancer surgery in hospitals worldwide, focusing on the effect of disease stage and complications on postoperative mortality. Methods: This was a multicentre, international prospective cohort study of consecutive adult patients undergoing surgery for primary breast, colorectal, or gastric cancer requiring a skin incision done under general or neuraxial anaesthesia. The primary outcome was death or major complication within 30 days of surgery. Multilevel logistic regression determined relationships within three-level nested models of patients within hospitals and countries. Hospital-level infrastructure effects were explored with three-way mediation analyses. This study was registered with ClinicalTrials.gov, NCT03471494. Findings: Between April 1, 2018, and Jan 31, 2019, we enrolled 15 958 patients from 428 hospitals in 82 countries (high income 9106 patients, 31 countries; upper-middle income 2721 patients, 23 countries; or lower-middle income 4131 patients, 28 countries). Patients in LMICs presented with more advanced disease compared with patients in high-income countries. 30-day mortality was higher for gastric cancer in low-income or lower-middle-income countries (adjusted odds ratio 3·72, 95% CI 1·70–8·16) and for colorectal cancer in low-income or lower-middle-income countries (4·59, 2·39–8·80) and upper-middle-income countries (2·06, 1·11–3·83). No difference in 30-day mortality was seen in breast cancer. The proportion of patients who died after a major complication was greatest in low-income or lower-middle-income countries (6·15, 3·26–11·59) and upper-middle-income countries (3·89, 2·08–7·29). Postoperative death after complications was partly explained by patient factors (60%) and partly by hospital or country (40%). The absence of consistently available postoperative care facilities was associated with seven to 10 more deaths per 100 major complications in LMICs. Cancer stage alone explained little of the early variation in mortality or postoperative complications. Interpretation: Higher levels of mortality after cancer surgery in LMICs was not fully explained by later presentation of disease. The capacity to rescue patients from surgical complications is a tangible opportunity for meaningful intervention. Early death after cancer surgery might be reduced by policies focusing on strengthening perioperative care systems to detect and intervene in common complications. Funding: National Institute for Health Research Global Health Research Unit
