Rational inattention in discrete choice models: Estimable specifications of RI-multinomial logit (RI-MNL) and RI-nested logit (RI-NL) models
Bibliographic record
Abstract
As opposed to the fully informed choice-making assumption in classical discrete choice models, the theory of Rational Inattention (RI) 1 in discrete choice modelling has been recently proposed in the literature. Matějka and McKay (2015) proposed the RI-multinomial logit (RI-MNL), and Fosgerau et al. (2020) proposed the RI-nested logit (RI-NL) model. These models consider that choice makers are bayesian agents with prior probabilities of choices and process any further information assuming an information processing cost to have the updated/posterior choice probabilities. However, the proposed RI-MNL and RI-NL models are theoretical formulations without any estimable empirical specifications. This paper proposes econometric formulations of RI-MNL and RI-NL models that are estimable using classical maximum likelihood estimation methods and suitable for revealed crossectional choice data. The proposed models are estimated for commuting mode choices in the Greater Toronto and Hamilton Area (GTHA) using data from a household travel survey conducted in the region. Empirical investigation reveals that the induction of RI in the classical discrete choice models (MNL and NL) improves the model fit by large margins. While scale parameterization in classical MNL and NL does not make a better model, the scale parameterization better captures the choice heterogeneity within the RI framework. Between the RI-MNL and RI-NL, the RI-NL is proven to be the best. The RI-NL model can capture asymmetric (between increasing and decreasing values) elasticities of choice attributes.
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 | 0.018 | 0.054 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.003 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.003 |
| Scholarly communication | 0.003 | 0.005 |
| Open science | 0.005 | 0.003 |
| Research integrity | 0.003 | 0.005 |
| Insufficient payload (model declined to judge) | 0.006 | 0.001 |
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".