Estimating Risk Relativity of Driving Records using Generalized Additive Models: A Statistical Approach for Auto Insurance Rate Regulation
Bibliographic record
Abstract
Abstract Studying driving records (DR) and assessing their risk relativity is crucial for auto insurance rate regulation. Typically, the evaluation of DR involves estimating risk using empirical loss cost or modeling approaches such as Generalized Linear Models (GLM). This article presents a novel methodology employing Generalized Additive Models (GAM) to estimate the risk relativity of DR. By treating the integer level of DR as a continuous variable, the proposed method offers enhanced flexibility and practicality in evaluating the associated risk. Extending the linear model to GAM is a critical advancement that harnesses advanced statistical methods in actuarial practice, providing a more statistically robust application of the proposed approach. Moreover, the integration of functional patterns with Class or Territory enables the investigation of statistical evidence supporting the existence of associations between risk factors. This approach helps address the issue of potential double penalties in insurance pricing and calls for a statistical solution to overcome this challenge. Our study demonstrates that utilizing the GAM approach yields a more balanced estimation of DR relativity, thereby reducing discrimination among different DR levels. This finding highlights the potential of this statistical method to improve fairness and accuracy in auto insurance rate making and regulation.
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 | 0.024 | 0.071 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.005 | 0.004 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".