Misspecified Model Estimation and Its Impact on Predictions

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: He, Junnan, Hu, Lin, Kovach, Matthew, Li, Anqi
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914470191693824
author He, Junnan
Hu, Lin
Kovach, Matthew
Li, Anqi
author_facet He, Junnan
Hu, Lin
Kovach, Matthew
Li, Anqi
contents We study a linear statistical model where outcomes depend on regressors with fixed population coefficients and observation-specific latent coefficients, along with measurement errors. A decision-maker estimates population coefficients and uses the estimates to predict the latent coefficients for a given observation. We analyze how misspecification of some population coefficients distorts predictions, investigating comparative statics with respect to: (1) residual information in regressors associated with misspecified coefficients after projecting out those associated with free coefficients, (2) alignment between misspecification vector and latent-to-coefficient mapping. Applications include employee rating with unconscious bias and LLM-mediated consumer research.
format Preprint
id arxiv_https___arxiv_org_abs_2309_08740
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Misspecified Model Estimation and Its Impact on Predictions
He, Junnan
Hu, Lin
Kovach, Matthew
Li, Anqi
Theoretical Economics
We study a linear statistical model where outcomes depend on regressors with fixed population coefficients and observation-specific latent coefficients, along with measurement errors. A decision-maker estimates population coefficients and uses the estimates to predict the latent coefficients for a given observation. We analyze how misspecification of some population coefficients distorts predictions, investigating comparative statics with respect to: (1) residual information in regressors associated with misspecified coefficients after projecting out those associated with free coefficients, (2) alignment between misspecification vector and latent-to-coefficient mapping. Applications include employee rating with unconscious bias and LLM-mediated consumer research.
title Misspecified Model Estimation and Its Impact on Predictions
topic Theoretical Economics
url https://arxiv.org/abs/2309.08740