compare.compind: Comparison of Composite Indices

Description Usage Arguments Details Value Author(s) References Examples

View source: R/compare.R

Description

Function which compares Composite Indices among entities

Usage

1
2
## S3 method for class 'compind'
compare(input, bench = NULL)

Arguments

input

calculated composite index: output of the function compind

bench

output of the function compind for the bechmark. Default to NULL, i.e. no benchmark used.

Details

- Compute the difference in the values of the composite indexes over time, and on average. - Compute the average shift in entities' ranks wrt a benchmark.

Value

A list with the following components:

IndexDifference: array of size (number of timeperiods x number of entities x number of entities) containing the pairwise enitity difference in the value of the composite index for a given time period.

AvIndexDifference: matrix of size (number of entities x number of entities) containing the time-average value of MeanIndexDifference.

AbsRankDifference: vector of size (number of time periods x 1) containing the absolue value of the average shift in entities' rank for a given time period. Is NULL if not benchmark is provided.

AvAbsRankDifference: timeperiod-average value of AbsRankDifference. Is NULL if no benchmark is provided.

AvRankCorrelation: timeperiod-average value of correlation between ranks. Is NULL if no benchmark is provided.

Author(s)

David Ardia, Kris Boudt, Valentin Todorov and Keshav Pratap

References

Saisana, M., Saltelli, A., Tarantola, S. (2005) Uncertainty and sensitivity analysis techniques as tools for the quality assessment of composite indicators. Journal of Royal Statistical Society 168, pp.307-323.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
  # Load data
  data(cip)
  # Compute Composite index (eight-indicators linear) for a range of years
  index_cip = compind(country~., time="year", for.period = 2005:2010, data=cip, ctr=ProcessControl(weights=c("fixed", 1/6, 1/6, 1/12, 1/12, 1/12, 1/12, 1/6, 1/6), aggregation="linear"))
  out = compare(index_cip)
  # Print and plot compare results
  print(out)
  plot(out)
  # Use a benchmark (a given year)
  bench= compind(country~., time="year", for.period = 2004, data=cip, ctr=ProcessControl(weights=c("fixed", 1/6, 1/6, 1/12, 1/12, 1/12, 1/12, 1/6, 1/6), aggregation="linear"))
  out = compare(index_cip, bench)
  print(out)

CItools documentation built on May 2, 2019, 4:48 p.m.