fptp2av: Simulate election outcomes for AV given FPTP votes.

In danielmarcelino/FPTP2AV: Simulate Election Outcomes for AV Given FPTP Votes

Description

This program simulates election outcomes for the Alternative Vote System using the actual First-Past-The-Post votes. It takes a wide table with the election results and returns a csv file with simulations for further analysis.

Usage

fptp2av(data = NULL, link = NULL)

Arguments

data	A flat table or data frame containing FPTP votes for each district.
link	A link function containing a structure for candidates/parties relations.

Value

A text file (csv) containing simulated electoral results for each district.

Note

Unfortunately, this function is not so flexible regarding the data supplied. For it to work properly, the input table should have the same format (flat table) as the example data provided within this package. Namely, the same structure with an Id column, the name of the constituencies in another column, and each party achievements distributed in columns. Besides the dataset, a "link" structure for party/candidates' relations is also necessary.

References

Marcelino, Daniel (2013). SciencesPo: A Tool Set for Analyzing Political Behaviour Data. Available at SSRN: http://dx.doi.org/10.2139/ssrn.2320547

Examples

#Feeding data:
data(ge2010)

# Let's take a small portion of the data.
IR <-ge2010[ge2010$Region=="Northern Ireland",]

# First, we create vectors for parties/candidates link
# For simplicity, I'm considering only parties which received some votes.

APNI<- c("SF", "DUP", "SDLP", "TUV", "UCUNF")
BNP<-c("Ch P", "Con", "CPA", "NF", "UKIP");
Ch_P<-c("BNP", "CPA", "Con", "ED")
Con<-c("LD", "UKIP", "ED", "Ch P", "CPA")
CPA<-c("BNP", "Ch P", "Con", "ED")
DUP<-c("SDLP", "TUV", "UCUNF", "APNI")
ED<-c("Con", "BNP", "NF", "CPA", "Ch P")
Grn<-c("LD", "Lab", "SDLP", "PC", "SSP", "TUSC", "TUV")
Lab<-c("LD", "Grn", "SDLP", "Soc", "SSP", "TUSC", "TUV")
LD<-c("Lab", "Con", "Grn", "PC", "SDLP")
MRLP<-c("Con", "Lab", "LD", "Grn", "PC", "SDLP", "SNP", "SSP", "UKIP")
NF<-c("BNP", "Con", "ED", "UKIP")
PC<-c("Con", "Lab", "LD")
Respect<-c("Con", "Lab", "LD")
SDLP<-c("Grn", "Lab", "LD")
SF<-c("APNI", "USUNF")
Soc<-c("Lab", "SSP", "SDLP", "TUSC", "TUV")
SNP<-c("Ch P", "CPA", "UKIP")
SSP<-c("Grn", "Lab", "LD", "SDLP", "TUSC", "TUV")
TUSC<-c("SDLP", "Soc", "SSP", "TUV")
TUV<-c("SDLP", "Soc", "SSP", "TUSC")
UCUNF<-c("APNI", "SF", "DUP")
UKIP<-c("BNP", "NF", "Con")

## vector with party/candidate relations:

party_chains <-structure(list(APNI,BNP,Ch_P,Con,CPA,DUP,ED,
Grn, Lab,LD,MRLP,NF,PC,Respect,SDLP,SF,Soc,SNP,SSP,TUSC,TUV,UCUNF,
 UKIP), .Names = c("APNI","BNP","Ch_P","Con","CPA","DUP","ED","Grn",
 "Lab","LD","MRLP","NF","PC","Respect","SDLP","SF","Soc","SNP","SSP",
 "TUSC", "TUV","UCUNF","UKIP" ));

## Run the algorithm
## Not run: fptp2av(data=IR, link=party_chains)

danielmarcelino/FPTP2AV documentation built on May 14, 2019, 4:05 p.m.