(1) log(A a B b )=a*log(A)+b*log(B)
(2) Increase each input (K, L & T) by l (so that, lK, lL &lT), output will rise proportionally: lY. Therefore, lY=(lK) a (lL) b (lT) g. Do the log-transformation. Show that “log(l)=(a+b+g)log(l)”, so that a+b+g=? (constant return to scale)
(3) Test the “constant return to scale” restriction derived in (2) (You may add an intercept in the regression)
(1) Try different combinations of P & Q (suggestion: 0, 1, 2). Use the AIC & SC to pick the best model(s). Check autocorrelation (of the error) in the selected model (e.g. LM test)
(3) Discuss the dynamics of the GDP growth of the two countries
… (i) build a suitable econometric model, (ii) run some simulations to predict the winner (based upon some inputs which you may assume), and (iii) write a short justification (2-3 hundred words) of your specific model and its results…
- Suitable model – something other than growth might affect vote too – try some of the given economic/political variables, justify why these variables might help to predict vote. Then, check if your conjecture is valid using empirical results (say, is the coefficient of the added variable significant? Any improvement in terms of adj-R 2 ?)
- Discuss your findings in 200-300 words…………….