Pre-Training Estimators for Structural Models: Application to Consumer Search
This approach makes structural models more accessible to researchers and practitioners by reducing computational cost and effort, though it is incremental as it builds on existing neural network methods.
The paper tackles the computational difficulty of estimating structural econometric models by introducing a pretrained estimator that uses a neural network to recognize model parameters from data patterns, achieving high accuracy and reducing estimation time to seconds on 12 real datasets.
We explore pretraining estimators for structural econometric models. The estimator is "pretrained" in the sense that the bulk of the computational cost and researcher effort occur during the construction of the estimator. Subsequent applications of the estimator to different datasets require little computational cost or researcher effort. The estimation leverages a neural net to recognize the structural model's parameter from data patterns. As an initial trial, this paper builds a pretrained estimator for a sequential search model that is known to be difficult to estimate. We evaluate the pretrained estimator on 12 real datasets. The estimation takes seconds to run and shows high accuracy. We provide the estimator at pnnehome.github.io. More generally, pretrained, off-the-shelf estimators can make structural models more accessible to researchers and practitioners.