EC 571 Advanced Econometrics

Homework 4

Part I

Based on a classical study of investment demand (Greene [2008], pp.250-252, Grunfeld and Griliches [1960], Boot and deWitt [1960]), the model is:

I_it = α_it + β_itF_it + γ_itC_it + ε_it

where
i = 10 firms: GM, CH, GE, WE, US, AF, DM, GY, UN, IBM.
t = 20 years: 1935-1954.
I_it = Gross investment.
F_it = Market value.
C_it = Value of the stock of plant and equipment.
e_it = Error term.

Data of above 3 variables for 10 companies are available in the previous homework.

Formulate and estimate the model as a system of regression equations with varying intercept and slop parameters across firms. That is,
α_it = α_i, β_it = β_i, γ_it = γ_i for all firm i and time t.
Formulate and estimate the model as a system of regression equations with varying intercept and slop parameters over time. That is,
α_it = α_t, β_it = β_t, γ_it = γ_t for all firm i and time t.

Part II

Comparing with the original specification of Klein Model I as presented in Greene [2011, Example 10.6 and Table 10.5], the alternative specification due to Goldberger (Econometric Theory, 1963, p.303) did not use any parameter restrictions. Instead, two more endowgenous variables and therefore two additional identities were used. The two extra endogenous variables are:

Y = National income (after tax income of both public and private sectors combined)
W = W1 + W2 (total wage income from both public and private sectors)

The Goldberger specification of the Klein Model I has the identical three stochastic equations as presented in Greene's. However, the five identities are:

X = Y + T - W2
Y + T = C + I + G
Y = P + W
W = W1 + W2
K = I + K(-1)

The variable G (government expenditure) in the Goldberger specification (as well as the original Klein specification) includes government employee's wage W2. However, the provided data series (and Greene's) of G does not include W2. Minor adjustment of the data series is needed. Formulate and estimate the Klein Model I based on Goldberger specification. The estimation results must be the same as those of the original specification.