Estimation of Random-Coefficient Demand Models: Two Empiricists' Perspective
By Christopher R. Knittel, Konstantinos Metaxoglou. Review of Economics and Statistics, 2014
153 citations recorded by the source corpus.
Abstract We document the numerical challenges we experienced estimating random-coefficient demand models as in Berry, Levinsohn, and Pakes (1995) using two well-known data sets and a thorough optimization design. The optimization algorithms often converge at points where the first- and second-order optimality conditions fail. There are also cases of convergence at local optima. On convergence, the variation in the values of the parameter estimates translates into variation in the models' economic predictions. Price elasticities and changes in consumer and producer welfare following hypothetical merger exercises vary at least by a factor of 2 and up to a factor of 5.
Subfields: Economics and Econometrics, Marketing
View Paper: https://www.semanticscholar.org/paper/f60c86200b9945984c718463afda1020a89c809c
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Economics and Econometrics
- D. Brunner
- A. Romahn
- Constantin Weiser
- Florian Heiss
- Chandra R. Bhat
- M. Polyakova
- Huseyin Topaloglu
- C. Michelsen
In Marketing
- D. Brunner
- A. Romahn
- Constantin Weiser
- Florian Heiss
- Chandra R. Bhat
- Huseyin Topaloglu
- Richard G. Newell
- Juha Siikamäki
Citations in the corpus
- Transparency, Reproducibility, and the Credibility of Economics Research (2018, unknown): “…some estimates from random-coefficient demand models, where problems with certain numerical maximization algorithms may produce misleading estimates (Knittel and Metaxoglou 2011; Knittel and Metaxoglou 2013); McCullough and Vinod (2003) contains a more general discussion of robustness and…”
- Willingness to Pay for Clean Air: Evidence from Air Purifier Markets in China (2020, unknown): “Kremer et al. (2011) uses a randomized control trial (RCT) for water pollution in Kenya.”
- Reliable estimation of random coefficient logit demand models (2017, methodology): “For the model’s exposition and to o↵er a meaningful comparison with the results in Knittel and Metaxoglou (2014), we assume that the K dimensions of ⌫ i are independently distributed.”