If we plug the γjs back into the expression for pj/k, we obtain,
_ exp(Yj) exp(β0zjk)
pj/k " PJ=1 exp(β0Zjk + Yj )
_ __________nj__exp(β0Zjt)
PK=1 exp(βzjk + Ik) PJ=1 exP(αj + βzjk)
_ exp(β0Zjk + Ik )
PtT=1 exp(β0zjk + Ik)
and the concentrated log-likelihood is that of a logit model where the choices
are now the sectors with an alternative specific constant added to the model.
This log-likelihood is equivalent to that of a Poisson regression with fixed-
effects (see, for example, Cameron & Trivedi (1998)].
24
More intriguing information
1. Valuing Access to our Public Lands: A Unique Public Good Pricing Experiment2. Weak and strong sustainability indicators, and regional environmental resources
3. The Provisions on Geographical Indications in the TRIPS Agreement
4. Elicited bid functions in (a)symmetric first-price auctions
5. An Incentive System for Salmonella Control in the Pork Supply Chain
6. Real Exchange Rate Misalignment: Prelude to Crisis?
7. The Mathematical Components of Engineering
8. Expectations, money, and the forecasting of inflation
9. On Dictatorship, Economic Development and Stability
10. CHANGING PRICES, CHANGING CIGARETTE CONSUMPTION