In: Economics
For the regression model be , which is the usual CD-production function, the linear transformation would be as or or , for Y be the natural log of GDP, k be the natural log of Fixed capital and l be the natural log of Employment.
The R-script for the regression estimation would be as below.
Year <- c(1955, 1956, 1957, 1958, 1959, 1960, 1961, 1962, 1963, 1964, 1965, 1966, 1967, 1968, 1969, 1970, 1971, 1972, 1973, 1974) GDP <- c(114043, 120410, 129187, 134705, 139960, 150511, 157897, 165286 ,178491, 199457, 212323, 226977, 241194, 260881, 277498, 296530, 306712, 329030, 354057 ,374977) Employment <- c(8310, 8529, 8738 ,8952 ,9171, 9569, 9527, 9662, 10334, 10981, 11746, 11521, 11540, 12066, 12297, 12955, 13338, 13738, 15924, 14154) FixedCapital <- c(182113, 193745, 205192, 215130, 225021, 237026, 248897, 260661, 275466, 295378, 315715, 337642, 363599, 391847, 422382, 455049, 484677, 520553, 561531, 609825) summary(lm(log(GDP) ~ log(FixedCapital) + log(Employment))) |
The output would be as below.
Call: Residuals: Coefficients: Residual standard error: 0.02829 on 17 degrees of freedom |
Hence, the estimated regression model would be as , ie .
Further solving, we have
or
ie , which is the required CD-production function.