Package 'semsfa'

Title: Semiparametric Estimation of Stochastic Frontier Models
Description: Semiparametric Estimation of Stochastic Frontier Models following a two step procedure: in the first step semiparametric or nonparametric regression techniques are used to relax parametric restrictions of the functional form representing technology and in the second step variance parameters are obtained by pseudolikelihood estimators or by method of moments.
Authors: Giancarlo Ferrara and Francesco Vidoli
Maintainer: Giancarlo Ferrara <[email protected]>
License: GPL
Version: 1.1
Built: 2024-12-22 06:49:40 UTC
Source: CRAN

Help Index


Semiparametric Stochastic Frontier Models

Description

Semiparametric Estimation of Stochastic Frontier Models following the two step procedure proposed by Fan et al (1996) and further developed by Vidoli and Ferrara (2015) and Ferrara and Vidoli (2017). In the first step semiparametric or nonparametric regression techniques are used to relax parametric restrictions regards the functional form of the frontier and in the second step variance parameters are obtained by pseudolikelihood or method of moments estimators. Monotonicity restrinctions can be imposed by means of P-splines.

Author(s)

Giancarlo Ferrara, Francesco Vidoli
Maintainer: Giancarlo Ferrara <[email protected]>

References

Aigner., D., Lovell, C.A.K., Schmidt, P., 1977. Formulation and estimation of stochastic frontier production function models. Journal of Econometrics 6:21-37

Fan, Y., Li, Q., Weersink, A., 1996. Semiparametric estimation of stochastic production frontier models. Journal of Business & Economic Statistics 14:460-468

Ferrara, G., Vidoli, F., 2017. Semiparametric stochastic frontier models: A generalized additive model approach. European Journal of Operational Research, 258:761-777.

Hastie, T., Tibshirani, R., 1990. Generalized additive models. Chapman & Hall

Kumbhakar, S.C., Lovell, C.A.K, 2000. Stochastic Frontier Analysis. Cambridge University Press, U.K

Meeusen, W., van den Broeck, J., 1977. Efficiency estimation from Cobb-Douglas production functions with composed error. International Economic Review, 18:435-444

Vidoli, F., Ferrara, G., 2015. Analyzing Italian citrus sector by semi-nonparametric frontier efficiency models. Empirical Economics, 49:641-658


Prediction of the individual efficiency score

Description

This function calculates and returns efficiency estimates from semiparametric stochastic frontier models estimated with semsfa().

Usage

efficiencies.semsfa(semobj, log.output = TRUE, ...)

Arguments

semobj

a stochastic frontier model object returned by semsfa()

log.output

logical. Is the dependent variable logged?

...

further arguments to the summary method are currently ignored

Details

The estimation of the individual efficiency score for a particular point (x,y)(x,y) on a production frontier might be obtained from the Jondrow et al. (1982) procedure. Defining:

σ2=σu2+σv2,u(x)=σu2ϵ/σ2,σ2=σu2σv2/σ2\sigma^2=\sigma_u^2+\sigma_v^2, u_{*}(x) = -\sigma_u^2 \epsilon/\sigma^2, \sigma_{*}^2=\sigma_u^2 \sigma_v^2/\sigma^2

it can be shown that:

uϵ N+(μ(x),σ2(x)).u|\epsilon ~ N^+(\mu_{*}(x),\sigma_{*}^{2}(x)).

We can use this distribution to obtain point previsions of uu trought the mean of the conditional distribution:

E(uϵ)=μ+σf(μ/σ)/(1F(μ/σ))E(u|\epsilon)=\mu_{*} + \sigma_{*} f(-\mu_{*}/\sigma_{*})/(1-F(\mu_{*}/\sigma_{*}))

where ff and FF represent the standard Normal density and cumulative distribution function, respectively; alternative formulas for cost frontier models are easy to get (please see Kumbhakar and Lovell, 2000).

If the response variable is measured in logs, a point estimate of the efficiency is then provided by exp(u)(0,1)\exp(-u) \in (0,1); otherwise, (fitt-u)/fitt where fitt is the estimated output evaluated at the frontier, given the inputs.

Value

An object of class semsfa containing the following additional results:

u

the prediction of the individual efficiency score

efficiencies

point estimate of the efficiency

Author(s)

Giancarlo Ferrara and Francesco Vidoli

References

Jondrow, J., Lovell, C.A.K., Materov, I.S., Schmidt, P., 1982. On the estimation of technical inefficiency in stochastic frontier production models. Journal of Econometrics 19, 233-238.

Kumbhakar, S.C., Lovell, C.A.K., 2000. Stochastic Frontier Analysis. Cambridge University Press, New York.

See Also

semsfa, summary.semsfa, plot.semsfa.

Examples

set.seed(0)
n<-200

#generate data
x<- runif(n, 1, 2)
fy<- 2+30*x-5*x^2
v<- rnorm(n, 0, 1)
u<- abs(rnorm(n,0,2.5))
#production frontier
y <- fy + v - u
dati<-data.frame(y,x)

#first-step: gam, second-step: fan (default)
o<-semsfa(y~s(x),dati,sem.method="gam")

#calculate efficiencies
a<-efficiencies.semsfa(o)

Pseudolikelihood estimator of the λ\lambda parameter

Description

Pseudolikelihood estimator of the λ\lambda parameter

Usage

fan(lambda_fan, resp, Ey, ineffD)

Arguments

lambda_fan

the λ=σu/σv\lambda=\sigma_u/\sigma_v parameter to be estimated

resp

the single response variable Y observed

Ey

the conditional expectation estimate obtained in the first step of the algorithm

ineffD

logical: TRUE for estimating a production function, FALSE for estimating a cost function; this is done for usage compatibility with frontier package

Value

Estimated λ\lambda parameter

Note

Internal usage only

Author(s)

Giancarlo Ferrara and Francesco Vidoli

References

Fan, Y., Li, Q., Weersink, A., 1996. Semiparametric estimation of stochastic production frontier models. Journal of Business & Economic Statistics 14:460-468


Default SEMSFA plotting

Description

This function plots the semiparametric/nonparametric intermediate model object estimated in the first step of the algorithm and, if efficiencies.semsfa() is esecuted, individual point estimate of the efficiency.

Usage

## S3 method for class 'semsfa'
plot(x, g.type, mod, ...)

Arguments

x

a semsfa object as returned from semsfa() or efficiencies.semsfa()

g.type

a character string indicating the type of plot. Possible values are: "reg" to plot the semiparametric/nonparametric model object estimated in the first step from semsfa(), "eff" to draw point estimate of the efficiency obtained from efficiencies.semsfa()

mod

a character string indicating the plot style for g.type="eff": "hist" for histogram and "dens" for density plot

...

further arguments passed to plot.default.

Value

The function simply generates plots.

Author(s)

Giancarlo Ferrara and Francesco Vidoli

See Also

semsfa, efficiencies.semsfa.

Examples

set.seed(0)
n<-200

#generate data
x<- runif(n, 1, 2)
fy<- 2+30*x-5*x^2
v<- rnorm(n, 0, 1)
u<- abs(rnorm(n,0,2.5))
#production frontier
y <- fy + v - u
dati<-data.frame(y,x)

#first-step: gam, second-step: fan (default)
o<-semsfa(y~s(x),dati,sem.method="gam")
#the following plot will be like that generated by plot.gam
plot(o,g.type="reg")

#adding a covariate
z<- runif(n, 1, 2)
dati$z<-z

#first-step: kernel, second-step: fan (default)
o<-semsfa(y~x+z,dati,sem.method="kernel")
#the plot will be like that generated by a plot.npreg
## Not run: plot(o,g.type="reg")

#calculate efficiencies ...
a<-efficiencies.semsfa(o)
plot(a,g.type="eff",mod="dens")

#adding further parameters as for plot.default: col, main, xlim, ...
plot(a,g.type="eff",mod="dens",col=2,main="Density Efficiency",xlim=c(0,1),xlab="Efficiency")

Semiparametric Estimation of Stochastic Frontier Models

Description

Semiparametric Estimation of Stochastic Frontier Models following the two step procedure proposed by Fan et al (1996) and further developed by Vidoli and Ferrara (2015) and Ferrara and Vidoli (2017). In the first step semiparametric or nonparametric regression techniques are used to relax parametric restrictions regards the functional form of the frontier and in the second step variance parameters are obtained by pseudolikelihood or method of moments estimators. Monotonicity restrinctions can be imposed by means of P-splines.

Usage

semsfa(formula, data = list(), sem.method = "gam", var.method = "fan",
       ineffDecrease=TRUE, tol = 1e-05, n.boot=0,...)

Arguments

formula

an object of class "formula": a symbolic description of the model to be fitted. The details of model specification are given under 'Details'

data

a data frame containing the variables in the model

sem.method

a character string indicating the type of estimation method to be used in the first step for the semiparametric or nonparametric regression; possible values are "gam" (default), "gam.mono" for monotone gam, "kernel" or "loess"

var.method

the type of estimation method to be used in the second step for the variance components: "fan" (default) for Fan et al. (1996) approach and "mm" for method of moments

ineffDecrease

logical: TRUE (default) for estimating a production function, FALSE for estimating a cost function; this is done for usage compatibility with frontier package

tol

numeric. Convergence tolerance for pseudolikelihood estimators of variance parameters of the composed error term

n.boot

numeric. Number of bootstrap replicates to calculate standard error for the variance components, by default bootstrap standard errors will not be calculated (n.boot=0)

...

further arguments accepted by mgcv::gam, gamlss::gamlss, np::npreg or loess

Details

Parametric stochastic production frontier models, introduced by Aigner et al. (1977) and Meeusen and van den Broeck (1977), specify output in terms of a response function and a composite error term. The composite error term consists of a two-sided error representing random effects and a one-sided term representing technical inefficiency. The production stochastic frontier model can be written, in general terms, as:

yi=f(xi)+viui,i=1,...,n,y_i = f (x_i)+v_i - u_i,\quad \quad i = 1, ..., n,

where YiR+Y_i\in R^+ is the single output of unit ii, XiRp+X_i\in R^{+}_{p} is the vector of inputs, f(.)f(.) defines a production frontier relationship between inputs X and the single output Y. In following common practice, we assume that vv and uu are each identically independently distributed (iidiid) with v N(0,σv)v~ N(0,\sigma_v) and uu distributed half-normally on the non-negative part of the real number line: u N+(0,σu)u~ N^{+}(0,\sigma_u); furthermore, the probability density function of the composite disturbance can be rewritten in terms of λ=σu/σv\lambda = \sigma_u/\sigma_v and σ2=σv2+σu2\sigma^2 = \sigma_v^2+\sigma_u^2 for the estimation algorithm. To overcome drawbacks due to the specification of a particular production function f()f(\cdot) we consider the estimation of a Semiparametric Stochastic Production Frontier Models through a two step procedure originally proposed by Fan et al (1996): in the first step a semiparametric or nonparametric regression technique is used to estimate the conditional expectation, while in the second step λ\lambda and σ\sigma parameters are estimated by pseudolikelihood (via optimize) or by method of moments estimators (var.method argument). In the case of a cost function frontier (ineffDecrease=FALSE) the composite error term is ϵ=v+u\epsilon = v + u. Vidoli and Ferrara (2015) suggest a Generalized Additive Model (GAM) framework in the first step even if any semiparametric or nonparametric tecnique may be used (Fan et al., 1996). The avalaible methods for the first step are:

  • sem.method="gam" invokes gam() from mgcv;

  • sem.method="gam.mono" invokes gamlss() from gamlss to impose monotonicity restrictions on inputs;

  • sem.method="kernel" invokes npreg() from np;

  • sem.method="loess" invokes loess() from stats.

Since in the first step different estimation procedure may be invoked from different packages, the formula argument has to be compatible with the corresponding function. The avalaible methods for the second step are:

  • var.method="fan" pseudolikelihood;

  • var.method="mm" Method of Moments.

Value

semsfa() returns an object of class semsfa. An semsfa object is a list containing the following components:

formula

the formula used

y

the response variable used as specified in formula

data

the data frame used

call

the matched call

sem.method

the type of semiparametric or nonparametric regression as given by sem.method ("gam", "gam.mono", "kernel", "loess")

var.method

the type of error component estimator ("fan", "mm")

ineffDecrease

logical, as given by ineffDecrease

reg

an object of class "gam", "gamlss" (monotone gam), "np"(kernel) or "loess" depending on sem.method

reg.fitted

fitted values on the "mean" frontier (semiparametric/non parametric regression)

regkewness

asymmetry index calculated on residuals obtained in the first step

lambda

λ\lambda estimate

sigma

σ\sigma estimate

fitted

fitted values on the frontier

tol

convergence tolerance for pseudolikelihood estimators used in optimize

residual.df

residual degree of freedom of the model

bic

'Bayesian Information Criterion' according to the formula -2*log-likelihood+ log(n)*npar where npar represents the number of parameters in the fitted model and n the number of observations

n.boot

number of bootstrap replicates used (default n.boot=0)

boot.mat

a matrix containing λ\lambda and σ\sigma values from each bootstrap replicate (if n.boot>0)

b.se

boostrapped standard errors for λ\lambda and σ\sigma (if n.boot>0)

Note

The function summary (i.e. summary.semsfa) can be used to obtain a summary of the results, efficiencies.semsfa to calculate efficiency scores and plot (i.e. plot.semsfa) to graph efficiency previsions and regression components (i.e. the first step).

You must take the natural logarithm of the response variable before fitting a stochastic frontier production or cost model.

Author(s)

Giancarlo Ferrara

References

Aigner., D., Lovell, C.A.K., Schmidt, P., 1977. Formulation and estimation of stochastic frontier production function models. Journal of Econometrics 6:21-37

Fan, Y., Li, Q., Weersink, A., 1996. Semiparametric estimation of stochastic production frontier models. Journal of Business & Economic Statistics 14:460-468

Ferrara, G., Vidoli, F., 2017. Semiparametric stochastic frontier models: A generalized additive model approach. European Journal of Operational Research, 258:761-777.

Hastie, T., Tibshirani, R., 1990. Generalized additive models. Chapman & Hall

Kumbhakar, S.C., Lovell, C.A.K, 2000. Stochastic Frontier Analysis. Cambridge University Press, U.K

Meeusen, W., van den Broeck, J., 1977. Efficiency estimation from Cobb-Douglas production functions with composed error. International Economic Review, 18:435-444

Vidoli, F., Ferrara, G., 2015. Analyzing Italian citrus sector by semi-nonparametric frontier efficiency models. Empirical Economics, 49:641-658

See Also

summary.semsfa, efficiencies.semsfa, plot.semsfa.

Examples

set.seed(0)
n<-200

x<- runif(n, 1, 2)
v<- rnorm(n, 0, 1)
u<- abs(rnorm(n,0,2.5))

#cost frontier
fy<- 2+30*x+5*x^2
y <- fy + v + u

dati<-data.frame(y,x)
#first-step: gam, second-step: fan
o<-semsfa(y~s(x),dati,sem.method="gam",ineffDecrease=FALSE)
#first-step: gam, second-step: mm
## Not run: o<-semsfa(y~s(x),dati,sem.method="gam",ineffDecrease=FALSE,var.method="mm")
plot(x,y)
curve(2+30*x+5*x^2,add=TRUE)
points(sort(x),o$fitted[order(x)],col=3,type="l")

#production frontier
fy<- 2+30*x-5*x^2
y <- fy + v - u

dati<-data.frame(y,x)
#first-step: gam, second-step: fan
o<-semsfa(y~s(x),dati,sem.method="gam",ineffDecrease=TRUE)

plot(x,y)
curve(2+30*x-5*x^2,add=TRUE)
points(sort(x),o$fitted[order(x)],col=3,type="l")

#imposing monotonicity restrictions on inputs
set.seed(25)
n=150

x=runif(n,0,3)
u=abs(rnorm(n,0,1))
v=rnorm(n,0,.75*((pi-2)/pi))

#production frontier
fy<-10-5*exp(-x)
y <- fy+v-u

dati<-data.frame(y,x)
#first-step: monotone gam, second-step: fan
o<-semsfa(y~pbm(x,mono="up"),sem.method = "gam.mono",dati)
plot(x,y)
curve(10-5*exp(-x),add=TRUE)
points(sort(x),o$fitted[order(x)],col=3,type="l")

Summary for semsfa object

Description

Create and print summary results of a stochastic frontier model object returned by semsfa() with regard to the "CONDITIONAL EXPECTATION ESTIMATE" of the first step and to the "VARIANCE COMPONENTS ESTIMATE" of the compound error.

Usage

## S3 method for class 'semsfa'
summary(object, ...)

Arguments

object

an semsfa object returned by semsfa()

...

further arguments to the summary method are currently ignored

Details

Please note that if bootstrap is carried out the tt-statistic is not reliable for testing the statistical significance of σ\sigma and λ\lambda, because these parameters are censored and cannot follow a tt-distribution. We suggest to compare the BIC of the semiparametric estimated model with the base model.

Value

summary.semsfa returns the summary of an object returned by semsfa() with few modifications if bootstrap is carried out:

b.t

tt-statistic given the bootstrapped standard errors for λ\lambda and σ\sigma (b.se)

b.pv

pp-values of the tt-statistic

Note

summary returns the same result if applied to an object created with semsfa or efficiencies.semsfa

Author(s)

Giancarlo Ferrara and Francesco Vidoli

See Also

semsfa, efficiencies.semsfa

Examples

#generate data
set.seed(0)
n<-200

x<- runif(n, 1, 2)
fy<- 2+30*x-5*x^2
v<- rnorm(n, 0, 1)
u<- abs(rnorm(n,0,2.5))
#production frontier
y <- fy + v - u

dati<-data.frame(y,x)

#first-step: gam, second-step: fan (default)
#without bootstrap
o<-semsfa(y~s(x),dati,sem.method="gam")
summary(o)

# ... with bootstrap
o<-semsfa(y~s(x),dati,sem.method="gam",n.boot=100)
summary(o)