plot.dpp_resolved: Plots a visual synthese of the errors induced by the...

In samthiriot/gosp.dpp: Pairing of entities to generate structured synthetic populations

Description

Plots all the values controlled by the algorithm, and compares their expected value (passed as data or parameter) and their actual value (as measured in the resulting synthetic population). The resulting graphs contain:

• a graph showing the relaxation parameters passed to the solving function, as computed by plot_relaxation

• a graph showing the Normalized Root Mean Square Error (NRMSE) for the each control variable, as computed by plot_errors

• a graph showing the population sizes asked for and generated for populations A and B, as computed by plot_population_sizes

• a graph showing the difference between the input and observed pairing probabilities pij, as computed by plot_errors_pij

• two graphs showing the initial and observed frequencies of control variables in both populations A and B, as computed by plot_frequencies_A

• two graphs showing the initial and observed average degrees in both populations A and B, as computed by plot_average_degree_A

• two graphs showing the difference between the expected and measured distribution of probability of degrees for both A and B, as computed by plot_errors_pdi

Usage

## S3 method for class 'dpp_resolved'
plot(x, nameA = "A", nameB = "B",
  colorRef = "darkgray", colorSynthetic = "blue", ...)

## S3 method for class 'dpp_result'
plot(x, nameA = "A", nameB = "B",
  colorRef = "darkgray", colorSynthetic = "blue", ...)

Arguments

x	an object returned by the matching.solve or the matching.generate method
nameA	a meaningfull label for the entity type of population A, such as "dwellings" (default to "A")
nameB	a meaningfull label for the entity type of population B, such as "households" (default to "B")
colorRef	the color to be used to plot values passed as parameters (defaults to "darkgray")
colorSynthetic	the color to be used to plot values measured in the synthetic population (defaults to "blue")
...	other parameters will be ignored quietly

Details

Note you can call the individual plotting functions for rendering for papers or zooming.

Value

nothing

Author(s)

Samuel Thiriot <[e-mail]>

Examples

data(dwellings_households)
prepared <- matching.prepare(
                     dwellings_households$sample.A, dwellings_households$sample.B, 
                     dwellings_households$pdi, dwellings_households$pdj, 
                     dwellings_households$pij)
solved <- matching.solve(
                     prepared, nA=50000, nB=40000, nu.A=1, phi.A=0, 
                     delta.A=0, gamma=0, delta.B=0, phi.B=0, nu.B=1, verbose=TRUE)
plot(solved, "dwellings", "households")
# we might as well drive the same on a generated case:
# sp <- matching.generate(
#                     solved, 
#                     sample.A=dwellings_households$sample.A, 
#                     sample.B=dwellings_households$sample.B, 
#                     verbose=TRUE)
# plot(sp, "dwellings", "households")

samthiriot/gosp.dpp documentation built on May 18, 2019, 3:44 p.m.