Please enable JavaScript.
Coggle requires JavaScript to display documents.
Nobel Prize winner: Heckman part 4 (Daniel McFadden (2000 NP: “for his…
Nobel Prize winner: Heckman part 4
Time series analysis: Granger and Engle
Granger: Cointegration
Two variables are both affected by a
third variable, which leads to a
misinterpretation of their relationship
Cointegration methods detect and correct this
Engle: Time varying volatility
Error terms not randomly distributed over time
Variables exhibit greater fluctuations in some periods than in others
--> ARCH-models to deal with this
Daniel McFadden
2000 NP: “for his development of theory and methods for
analyzing discrete choice” (joint with James Heckman).
Contribution:
Showed how to statistically handle fundamental aspects of microdata
Discrete choice models in microeconometrics
Discrete choice analysis McFadden
Discrete choice models: regression methods that take into account the binary nature of a variable of interest
Econometrics evolving
Deductive approach
Tinbergen
Haavelmo: “Econometrics has to be founded on theories that describe in a reasonably accurate way the fashion in which the observed world has operated in the past.”
Frisch: “Facts that speak for themselves speak in a very naïve language.”
Inductive approach
MacFadden
Heckman
Conclusion
Econometrics has developed enormously since 1920s
Availability computers
Data availability
Mathematical formulas allowed empirical testing of theories
New techniques replacing old