README.md

IndexNumR

Build
Status

Codecov test
coverage

CRAN

An R package for computation of index numbers

IndexNumR provides a set of functions for computing various bilateral and multilateral indices. It is designed to compute price or quantity indices over time. Bilateral indices include Laspeyres, Paasche, Fisher, Tornqvist, Sato-Vartia, Walsh, CES, geometric Laspeyres, geometric Paasche, time-product-dummy and Geary-Khamis as well as elementary indices Dutot, Carli, Harmonic mean, CSWD and Jevons. All of these bilateral indices can be computed as period-on-period, fixed-base or chained.

Multilateral indices can be computed in the time series context using the GEKS, Weighted Time Product Dummy or Geary-Khamis indexes, and updating is provided via the window, movement, half, mean, fbew or fbmw splicing methods. The GEKS method is computed using either the Fisher, Tornqvist, Walsh, Jevons or TPD index number methods.

The package also provides functions to compute measures of dissimilarity between time periods, which can be used to choose the linking period for chained indices.

A sample dataset is included called CES_sigma_2, which is a dataset containing prices and quantities on four products over twelve periods, but additional datasets assuming different values of the elasticity of substitution can be computed using the function CESData.

There is also a function dominicksData that can be used to download the Dominicks Finer Foods scanner data from the Chicago Booth School of Business.

Installation

You can install the released version of IndexNumR from CRAN with:

install.packages("IndexNumR")

And the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("grahamjwhite/IndexNumR")

Example

To estimate a simple chained Laspeyres price index using the CES_sigma_2 dataset,

library(IndexNumR)

priceIndex(CES_sigma_2, 
           pvar = "prices", 
           qvar = "quantities", 
           pervar = "time", 
           prodID = "prodID", 
           indexMethod = "laspeyres", 
           output = "chained")
#>            [,1]
#>  [1,] 1.0000000
#>  [2,] 0.9673077
#>  [3,] 1.2905504
#>  [4,] 1.3382002
#>  [5,] 1.2482444
#>  [6,] 1.7346552
#>  [7,] 1.6530619
#>  [8,] 1.4524186
#>  [9,] 1.8386215
#> [10,] 1.7126802
#> [11,] 2.1810170
#> [12,] 2.2000474

A GEKS index with mean splicing and an 11 period window is estimated as follows,

GEKSIndex(CES_sigma_2, 
          pvar = "prices", 
          qvar = "quantities", 
          pervar = "time", 
          prodID = "prodID", 
          indexMethod = "tornqvist", 
          window = 11, 
          splice = "mean")
#>            [,1]
#>  [1,] 1.0000000
#>  [2,] 0.8927314
#>  [3,] 1.0776386
#>  [4,] 1.1127724
#>  [5,] 0.9310834
#>  [6,] 1.1785361
#>  [7,] 1.1219447
#>  [8,] 0.9380228
#>  [9,] 1.0951667
#> [10,] 0.9501914
#> [11,] 1.1277725
#> [12,] 1.1330748

More examples are contained in the package vignette.

vignette("indexnumr", package = "IndexNumR")


Try the IndexNumR package in your browser

Any scripts or data that you put into this service are public.

IndexNumR documentation built on July 21, 2021, 5:08 p.m.