Abstract
Fang and Wang's (2015) Proposition 2 claims generic identification of a dynamic discrete choice model with hyperbolic discounting under exclusion restrictions. We note that Proposition 2 uses a definition of “generic” that does not preclude that a generically identified model is nowhere identified. We provide two examples of models that are generically identified under this definition, but that are, respectively, everywhere and nowhere identified. We then show that the proof of Proposition 2 is incorrect and incomplete. We conclude that Proposition 2 has no implications for identification of the dynamic discrete choice model and suggest alternative approaches to its identification.
Original language | English |
---|---|
Pages (from-to) | 565-571 |
Journal | International Economic Review |
Volume | 61 |
Issue number | 2 |
DOIs | |
Publication status | Published - May 2020 |