bquast / gvc Goto Github PK

View Code? Open in Web Editor NEW

19.0 6.0 9.0 454 KB

Global Value Chain tools

Home Page: http://qua.st/gvc

R 100.00%

gvc's Introduction

gvc

Several tools for Global Value Chain ('GVC') analysis are implemented

'Importing to Export' (i2e())
'Exporting to Re-export' (e2r())
'New Revealed Comparative Advantage' (nrca())
'Upsteamness' (upstream())
'Downstreamness' (downstream())
'Domestic Final Demand Foreign Value Added' (dfdfva())
'Foreign Final Demand Domestic Value Added' (ffddva())
'Domestic Final Demand Domestic Value Added' (dfddva())

Demonstration

Installation

The gvc package is available on CRAN, please note that this is a beta version, the API will change.

install.packages('gvc')

Additionally the development version, to be used at your own peril, is available from GitHub. Installation from GitHub is done using:

if (!require('remotes')) install.packages('remotes')
remotes::install_github('bquast/gvc')

Usage

Following installation, the package can be loaded using:

library(gvc)

For information on using the package, please refer to the help files.

help('gvc')
help(package = 'gvc')

For examples of usage, see the function specific help pages.

help('i2e')
help('e2r')
help('nrca')
help('upsteam')
help('downstream')
help('dfdfva')
help('ffddva')
help('dfddva')

In addition to the help files we provide long form examples in the vignette.

vignette('gvc')

# or
browseVignettes('gvc')

Additional Information

An overview of the changes is available in the NEWS file.

news(package='gvc')

There is also a blog post with information on my personal website.

https://qua.st/gvc/

This package relies on the diagonals package (CRAN).

Development

Development takes place on the GitHub page.

https://github.com/bquast/gvc

Bugs can be filed on the GitHub issues page.

https://github.com/bquast/gvc/issues

gvc's People

Contributors

Stargazers

Watchers

Forkers

memed01 desval brancengregory nishimaomaoxiong tchristina amrofi sanguovbobo anhnguyendepocen zzy6zzy

gvc's Issues

Output of i2e function

It seems like some of the output of the i2e command produces a false output. In the case of Sweden, one receives values of up to 12825, which, given the underlying concept of a VAX ratio or vertical specialisation ratios in general, seems unrealistic.
Attached, the output generated by i2e for the year 2011.

matrix A in calculating downstreamness & upstreamness

Hello, I use your package in R. Why when calculating the downstreamness and upstreamness, the matrix of coefficients is transposed? Why are domestic entries diagonally set to 0?

In the literature (https://unstats.un.org/unsd/trade/events/2016/newyork-egm/documents/background/OECD%20-%202013%20-%20GVC%20Mappings.pdf), the formula for downstreamness is

N = U* (I-A)^-1, where I is an identity matrix and A is the matrix of technical coefficients

for upstreamness is

D = U* (I-G)^-1 , where I is the identity matrix and G a matrix of output coefficients

NaN in upstream result

Hello!
In the results of the upstream calculation, all the values of NaN, while there were no errors. Downstream is calculated and does not have a similar error. What could be the reason?

Report a bug to make the result of the 'dfddva' function incorrect

A part of 'dfddva' function in gvc package needs correct:

dfddva <- function (x, aggregate = FALSE) {

    x <- data.frame(Importing_Country = rep(k, each = N), 
        Source_Industry = rep(i, times = G), **dfdfva** = x)
    return(x)
}

}

The result of the 'dfddva' function has incorrect name.
It means that 'dfdfva' =x needs to be replaced with 'dfddva'=x in the parentheses of the data.frame.

matrix A in downstream() & upstream()

Hi,

both the downstream and upstream functions make use of the matrix A, e.g.

downstream = function (x) { solve(diag(x$GN) - t(x$A)) %*% matrix(1, nrow = x$GN) }

where x refers to an object of class "decompr" as created using the load_tables_vectors() function from the decompr package.
However, the load_tables_vectors() does not save an object of name A in its output. It only saves the object Am, which is the "Imported / Exported goods IO shares matrix (x column-normalized by output o, with domestic entries set to 0)". When indexing the decompr object x with x$A, it is thus actually x$Am that is extracted (due to partial indexing).

Was this done deliberately or does the downstream function perhaps intend to use the A matrix instead?

Thanks.