PreprintarXiv licencearXiv
Does Calibration Pick the Right Model? Likelihood Evidence from Income Risk
Abstract
Quantitative macroeconomics typically disciplines models by matching selected moments. We show how finite-state hidden Markov approximations make likelihood estimation and model selection feasible for nonlinear, non-Gaussian latent processes on short panels, without simulation. We illustrate the method with PSID earnings data. The calibrations that best match moments of earnings dynamics fit complete histories worse than a likelihood-estimated process with a Gaussian persistent component. In a standard saving model, calibrations of one process that fit the targets about equally well imply mean assets that differ by a factor of three. Close agreement with selected moments can leave economic predictions largely undetermined.
Related
- Scalable, Likelihood-Free Calibration of Ice-Sheet Models with Deep Diffusion Emulators and Feature Matching
- Copula Geometry and Second-Order Calibration of Heavily Right Aggregation
- When Does Trimming Help Conformal Prediction? A Retained-Law Diagnostic under Calibration Contamination
- Calibration Count Reuse: Validity Does Not Determine Efficiency
- When does conformal calibration need censoring weights? Cause-of-failure prediction sets under competing risks
BibTeX
@misc{cheng2026does,
title = {{Does Calibration Pick the Right Model? Likelihood Evidence from Income Risk}},
author = {Jiuyi Cheng and Alexis Akira Toda},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2611.02029},
eprint = {2611.02029},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.02029},
}
RIS
TY - UNPB
TI - Does Calibration Pick the Right Model? Likelihood Evidence from Income Risk
AU - Jiuyi Cheng
AU - Alexis Akira Toda
PY - 2026
DA - 2026/11/01
PB - arXiv
DO - 10.48550/arxiv.2611.02029
UR - https://arxiv.org/abs/2611.02029
AB - Quantitative macroeconomics typically disciplines models by matching selected moments. We show how finite-state hidden Markov approximations make likelihood estimation and model selection feasible for nonlinear, non-Gaussian latent processes on short panels, without simulation. We illustrate the method with PSID earnings data. The calibrations that best match moments of earnings dynamics fit complete histories worse than a likelihood-estimated process with a Gaussian persistent component. In a standard saving model, calibrations of one process that fit the targets about equally well imply mean assets that differ by a factor of three. Close agreement with selected moments can leave economic predictions largely undetermined.
ER -