Essays in Econometrics, Industrial Organization, and Technology
Bibliographic record
Abstract
Recent developments in machine learning (ML) have facilitated the development of new methodologies to estimate structural economic models. This thesis contributes to the existing literature by developing new methodologies that combine Bayesian statistics with ML for structural model estimation and inference. In Chapter 1, I develop a model of the short-term accommodations market in Austin, Texas on Airbnb using Bayesian Neural Networks and simulate the effect of allowing platform fees to vary with consumer demand. I find that scaling fees with demand could have earned Airbnb an additional 10% in revenue over the period 2015-2019. In Chapter 2, I develop a method to estimate a first price sealed bid auction model using less observable data than what has conventionally been considered necessary. By combining a Bernstein polynomial with a Metropolis-Hastings sampler, I obtain point estimates and credible intervals on all primitives. Chapter 3 is joint work with my supervisor Martin Burda. We develop a statistical method to impose functional shape constraints on functions modeled by a flexible neural network that incorporates uncertainty inherent in the network structure. Further, we develop a method of architectural learning which enables simultaneous estimation via MCMC of both the network parameters and architecture.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.013 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.005 |
| Science and technology studies | 0.001 | 0.004 |
| Scholarly communication | 0.004 | 0.003 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.002 | 0.004 |
| Insufficient payload (model declined to judge) | 0.017 | 0.005 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".