error_lphom: Global error of a lphom estimated table

In lphom: Ecological Inference by Linear Programming under Homogeneity

Global error of a lphom estimated table

Description

Estimation of the error index (EI) of a RxC vote transfer matrix obtained with lphom()

Usage

error_lphom(
  lphom.object,
  upper.alfa = 0.1,
  show.plot = TRUE,
  num.d = 11,
  B = 30
)

Arguments

lphom.object	An object output of the lphom() function.
upper.alfa	Upper bound that will not be exceed by the EI estimate with a confidence 1 - alpha. By default, 0.10.
show.plot	TRUE/FALSE. Indicates whether the plot showing the relationship between EI and HETe estimated by simulation for the election under study should be displayed as a side-effect. By default, TRUE.
num.d	Number maximum of different disturbances, d, to be initially considered. Positive integer greater than or equal to 5. By default, 11.
B	Integer that determines the number of simulations to be performed for each disturbance value. By default, 30.

Value

A list with the following components

EI.estimate	Point estimate for EI.
EI.upper	Upper bound with confidence 1 - alpha of the EI estimate
figure	ggplot2 object describing the graphical representation of the relationship between EI and HETe.
equation	lm object of the adjustment between EI and HETe.
statistics	A four column matrix with the values of HET, HETe, EI and d associated with each simulated scenario.
TMs.real	Array with the simulated real transfer matrices associated with each scenario.
TMs.estimate	Array with the estimated transfer matrices associated with each scenario.

Note

ggplot2 is needed to be installed for this function to work.

See equation (12) in Romero et al. (2020) for a definition of the EI index.

Author(s)

Jose M. Pavia, [e-mail]

Rafael Romero [e-mail]

References

Romero, R, Pavia, JM, Martin, J and Romero G (2020). Assessing uncertainty of voter transitions estimated from aggregated data. Application to the 2017 French presidential election. Journal of Applied Statistics, 47(13-15), 2711-2736. doi: 10.1080/02664763.2020.1804842

See Also

lphom confidence_intervals_pjk

Examples

mt.lphom <- lphom(France2017P[, 1:8], France2017P[, 9:12], "raw", NULL, FALSE)
set.seed(253443)
example <- error_lphom(mt.lphom, upper.alfa = 0.10, show.plot = FALSE, num.d = 5, B = 8)
example$EI.estimate

lphom documentation built on March 21, 2022, 9:09 a.m.