Estimating Search with Learning
In this paper we estimate a structural model of search fordifferentiated products, using a unique dataset of consumer onlinesearch for hotels. We propose and implement an identification strategythat allows us to separately estimate consumer's beliefs, search costsand preferences. Learning plays an essential role in this strategy. Itcreates variation of posterior beliefs across consumers that isorthogonal to the variation in search costs. We show that ignoringendogeneity of choice sets due to search may lead to significant biasesin estimates of consumer demand: from 50 to more than 200 percentdepending on informational assumptions. Second, th median search costis about 25 dollars per 15 hotels; there is also a significantheterogeneity of search costs among the population. We perform astatistical test between models of search from known (Stigler 1967) andfrom unknown (Rothschild 1974) distribution and find that our datafavors the latter: we find a statistically significant amount ofBayesian learning.
Year of publication: |
2008
|
---|---|
Authors: | Koulayev, Sergei |
Institutions: | Columbia University |
Saved in:
freely available
Saved in favorites
Similar items by person
-
Explaining adoption and use of payment instruments by U.S. consumers
Koulayev, Sergei, (2012)
-
Sustained credit card borrowing
Grodzicki, Daniel, (2021)
-
Estimating demand in search markets: The case of online hotel bookings
Koulayev, Sergei, (2009)
- More ...