Comments on Estimation of Quasi-Hyperbolic Discount Function Using Observational Discrete Choice Data
- 주제(키워드) time preferences , quasi-hyperbolic discounting , dynamic discrete choice models
- 주제(기타) 경제학
- 설명문(URI) https://www.kci.go.kr/kciportal/ci/sereArticleSearch/ciSereArtiView.kci?sereArticleSearchBean.artiId=ART002703886
- 등재 KCI등재
- 발행기관 한국산업조직학회
- 발행년도 2021
- URI http://www.dcollection.net/handler/ewha/000000181200
- 본문언어 영어
초록/요약
In the literature on dynamically inconsistent preferences, there is a concern over the feasibility of estimating time preference parameters using observational data. This paper examines the estimability of time preference parameters by performing Monte Carlo simulations of the time preference estimator proposed in the dynamic discrete choice literature. The analysis shows that precise estimation of the quasi-hyperbolic discount function based on observational discrete choice data is indeed challenging and requires a very large sample. The analysis also shows that wrongly assuming exponential discounting, when the true time preferences are quasi-hyperbolic, could lead to an estimate of discount rate that is even smaller than the true long-run discount rate, let alone the true short-run discount rate, implying that researchers cannot be assured that they will at least get things right ‘on average.’ I further discuss other estimation issues empirical researchers are likely to encounter in estimating quasi-hyperbolic time preferences.
more