Displaying 3 results from an estimated 3 matches for "battese".
Did you mean:
attese
2008 Nov 08
0
New package "frontier" for Stochastic Frontier Analysis (SFA)
...Hence, the R package "frontier"
should have the same capabilities as "Frontier 4.1", i.e. Maximum Likelihood
Estimation of Stochastic Frontier Production and Cost Functions. Two
specifications are available: the error components specification with
time-varying efficiencies (Battese and Coelli, 1992) [5] and a model
specification in which the firm effects are directly influenced by a number
of variables (Battese and Coelli, 1995) [6].
The functions "front41WriteInput" and "front41ReadOutput" for creating
input files for and reading output files of &...
2008 Nov 08
0
New package "frontier" for Stochastic Frontier Analysis (SFA)
...Hence, the R package "frontier"
should have the same capabilities as "Frontier 4.1", i.e. Maximum Likelihood
Estimation of Stochastic Frontier Production and Cost Functions. Two
specifications are available: the error components specification with
time-varying efficiencies (Battese and Coelli, 1992) [5] and a model
specification in which the firm effects are directly influenced by a number
of variables (Battese and Coelli, 1995) [6].
The functions "front41WriteInput" and "front41ReadOutput" for creating
input files for and reading output files of &...
2008 Jun 16
1
Error in maximum likelihood estimation.
...r
loglik <- (-0.5 * nrow(x) *(log(2*pi) + log(sigma2))
-0.5 * sum(( y - x %*% beta + z %*% delta)^2/sigma2)
-sum(log(pnorm(d))) + sum(log(pnorm(dstar))))
return(-loglik)
}
-----------------------------------------------
Loglikelihood function is from page 21of Battese and Coelli (1993).
(You can download this article at http://www.une.edu.au/economics/publications/econometrics/emwp69.PDF
)
To test the above function with an artificial data set, I created the
following data.frame.
-----------------------------------------------
x1 <- abs(rnorm(100))*1...