A Seamless Integration of Gretl and R"

In gretlR: A Seamless Integration of 'Gretl' and 'R'

knitr::opts_chunk$set(echo = TRUE,comment = NULL,eval = F)
if(!require(gretlR)) install.packages('gretlR')
if(!require(kableExtra)) install.packages('kableExtra')
options(knitr.kable.NA = '')
library(gretlR)

About gretlR

gretlR is an R package that can run gretl program from R Markdown.

Installation

gretlR can be installed using the following commands in R.

install.packages("gretlR")

          OR

devtools::install_github('sagirumati/gretlR')

Usage

Please load the gretlR package as follows:

```r `r ''`                                                            
library(gretlR)
```

Then create a chunk for gretl as shown below:

```{gretl gretlR,eval=T,echo=T,comment=NULL} `r ''`
nulldata 500
set seed 13
gretl1 = normal()
gretl2 = normal()
setobs 12 1980:01 --time-series
gnuplot gretl1 --time-series --with-lines --output="line.png"
gnuplot gretl2 gretl1 --output="scatter.png"
ols gretl1 const gretl2
modeltab add
tabprint --output="olsTable.Rmd"
tabprint --output="olsTable.tex"
tabprint --output="olsTable.csv"
eqnprint --output="olsEquation.tex"
```  

```{gretl gretlR,eval=T,echo=F,comment=NULL} nulldata 500 set seed 13 gretl1 = normal() gretl2 = normal() setobs 12 1980:01 --time-series gnuplot gretl1 --time-series --with-lines --output="line.png" gnuplot gretl2 gretl1 --output="scatter.png" ols gretl1 const gretl2 modeltab add tabprint --output="olsTable.Rmd" tabprint --output="olsTable.tex" tabprint --output="olsTable.csv" eqnprint --output="olsEquation.tex"

The above chunk creates a gretl program with the chunk’s content, then
automatically run the gretl script, which will save gretl outputs in the new folder `gretlR` created in the current  working directory.

# include_graph function

We can *dynamically and reproducibly* fetch the `gretl` graph object we created with the `gretl` chunk using the following R chunk: 

For the scatter graph:

```r

include_graph(chunk = "gretlR",graph = "scatter.png")

or the line graph:

include_graph(chunk = "gretlR",graph = "line.png")

include_tex function

we can also include the equation of the OLS generated by the gretl chunk and save as olsEquation.tex.

If the output is pdf, one can use the raw LaTeX codes as follows:

\input{gretlR/gretlR/olsEquation.tex} \input{gretlR/gretlR/olsEquation.tex}

Or use include_tex function to include the equation as shown below:

include_tex(chunk = "gretlR",tex = "olsEquation")

include_tex(chunk = "gretlR",tex = "olsTable",start = 7,end = 24)

The OLS table output is saved by the gretl chunk as olsTable.Rmd. The entire OLS table output can included as child document as follows:

```r `r ''`

```

import_kable function

The gretl chunk also saves the OSL table as olsTable.csv. The import_kable function can be used to import it as a table. further customisation can be done with kableExtra package.

import_kable(chunk = "gretlR",file = "olsTable.csv",caption="Table generated from gretl 
             chunk", start=3,end=7,digits=2) %>% 
kableExtra::kable_styling(latex_options = c("basic","hold_position","scale_down")) %>% 
 kableExtra::row_spec(0,bold=T)

write_inp function

This function writes gretl file.

code=r'(nulldata 500
set seed 13
gretl1 = normal()
gretl2 = normal()
setobs 12 1980:01 --time-series
gnuplot gretl1 --time-series --with-lines --output="line.png"
gnuplot gretl2 gretl1 --output="scatter.png"
)'

write_inp(code,path="gretlCodes")

exec_inp function

This function executes existing gretl files.

code=r'(nulldata 500
set seed 13
gretl1 = normal()
gretl2 = normal()
setobs 12 1980:01 --time-series
gnuplot gretl1 --time-series --with-lines --output="line.png"
gnuplot gretl2 gretl1 --output="scatter.png"
 )'
write_inp(code,path="SomeFolder/gretlCodes")
exec_inp("someFolder/gretlCodes")

exec_gretl function

This function creates gretlfile from R object or a set of character strings and executes it. It is a combination of write_inp and exec_inp functions.

code=r'(nulldata 500
set seed 13
gretl1 = normal()
gretl2 = normal()
setobs 12 1980:01 --time-series
gnuplot gretl1 --time-series --with-lines --output="line.png"
gnuplot gretl2 gretl1 --output="scatter.png"
 )'
exec_gretl(code)

Demo

Demo can be accessed via demo(package="gretlR").

demo(exec_inp) 
demo(write_inp)
demo(exec_gretl)

R Markdown template

The R Markdown template for the gretlR can be accessed via file -> New File -> R Markdown -> From Template -> gretlR

Please download a set of example files from Github.

gretlR documentation built on May 1, 2022, 5:06 p.m.