Comparisons of Lagged Local Polynomial Regression COVID-19 Models
Bibliographic record
Abstract
One of the key challenges hindering accurate statistical modeling of infectious COVID-19 data is the dynamic nature of the evolving virus variants. Traditional statistical models typically assume key features of a phenomenon (such as associations between infection and mortality rates of a virus) remain constant, which is not the case with many infectious diseases. Motivated by the need for accurate predictions during the recent COVID-19 pandemic, this thesis considers a novel modeling approach: lagged time-varying coefficient regression. These models are a powerful tool which accounts for dynamic relationships among variables. This thesis builds upon the work of Liu et al. (2023), who demonstrate that lagged time-varying coefficient models yield promising results, especially for COVID-19 time series with complex relationships between reported case and death counts. However, the aforementioned paper does not exhaustively explore the impact of using different methods for bandwidth selection (a key component of time-varying coefficient regression) nor the effectiveness of these models on different types of time series (cumulative vs. daily) and different lengths of time series. For this reason, this thesis investigates several enhancements Liu et al.’s models, including assessing global bandwidth selection procedures, the effectiveness of using different data types and the impact of different time series lengths. An R implementation for these models is provided.
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 | 0.013 | 0.032 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.003 | 0.001 |
| Research integrity | 0.001 | 0.003 |
| Insufficient payload (model declined to judge) | 0.010 | 0.002 |
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".