Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/JanMarvin/nlsur
R package for the estimation of nonlinear least squares for equation systems
https://github.com/JanMarvin/nlsur
equation-systems fgnls ifgnls nls r r-package seemingly-unrelated-regression
Last synced: about 2 months ago
JSON representation
R package for the estimation of nonlinear least squares for equation systems
- Host: GitHub
- URL: https://github.com/JanMarvin/nlsur
- Owner: JanMarvin
- License: other
- Created: 2017-08-02T16:15:28.000Z (about 7 years ago)
- Default Branch: main
- Last Pushed: 2024-06-22T07:01:15.000Z (3 months ago)
- Last Synced: 2024-06-23T17:15:58.384Z (3 months ago)
- Topics: equation-systems, fgnls, ifgnls, nls, r, r-package, seemingly-unrelated-regression
- Language: R
- Homepage:
- Size: 4.47 MB
- Stars: 14
- Watchers: 5
- Forks: 4
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# NLSUR 
`nlsur` is a package to estimate a nonlinear least squares for single equations
or systems of equations.
The function to interact with is `nlsur()`. This function can estimate Nonlinear-Least Squares (NLS), Feasible Generalized NLS (FGNLS) and Iterative FGNLS (IFGNLS).
The packages supports a variety of functions like `print()`, `coef()`, `summary()`, `logLik()`, `vcov()` and `predict()`.
## Installation
With `drat`:
```R
drat::addRepo("JanMarvin")
install.packages("nlsur")
```
With `devtools`:
```R
devtools::install_github("JanMarvin/nlsur")
```
## Application
With `nlsur()` it is rather straight forward to estimate nonlinear demand systems. As example the following Translog demand system can be estimated.
```R
data(costs)
dat <- costs
# apply a patch to create Greene Ed. 7 Data
dat$Sm[dat$Year == 1958] <- 0.61886
dat$Pe[dat$Year == 1950] <- 1.12442
dat$Pm[dat$Year == 1949] <- 1.06625
# model
model <- list(
Sk ~ bk + dkk * log(Pk/Pm) + dkl * log(Pl/Pm) + dke * log(Pe/Pm),
Sl ~ bl + dkl * log(Pk/Pm) + dll * log(Pl/Pm) + dle * log(Pe/Pm),
Se ~ be + dke * log(Pk/Pm) + dle * log(Pl/Pm) + dee * log(Pe/Pm)
)
erg <- nlsur(eqns = model, data = dat, type = "FGNLS")
erg
```
Additional parameters may be obtained using `nlcom()` a wrapper for the delta method.
```R
# indirect estimation of translog parameters
bm <- nlcom(object = erg, form = "1 -be -bk -bl", rname = "bm")
dkm <- nlcom(object = erg, form = "-dkk -dkl -dke", rname = "dkm")
dlm <- nlcom(object = erg, form = "-dkl -dll -dle", rname = "dlm")
dem <- nlcom(object = erg, form = "-dke -dle -dee", rname = "dem")
# and now dmm (nlcom can search for parameters)
dmm <- nlcom(object = erg, form = "-dkm -dlm -dem", rname = "dmm")
```