Vintage Analysis in R

vintager transforms source data stored in PostgreSQL or Oracle database into vintage analysis format. From detailed list of units (e.g. loans) and events (e.g. repayment), vintage curves are calculated.

Pre-requisites

R version 3.02 or higher
PostgreSQL version 9.1 or higher
ggplot2, sqldf, RPostgreSQL, devtools, XLSWrite, reshape2

Installation

Package installation

library(devtools)
install_github("vintager", username="tomasgreif")
library(vintager)

Create time_distance function and load sample data

drv <- dbDriver("PostgreSQL")
# Use real values!
pgCon <- dbConnect(drv, user = 'usr', dbname="db", password = 'secret',
                   host = 'localhost', port = 5432)    
createTimeDistanceFunction(connection = con, loadData = TRUE)

Usage

vintager can work with the following data design:

TBD

In the basic form, you can just use:

drv <- dbDriver("PostgreSQL")
# Use real values!
pgCon <- dbConnect(drv, user = 'usr', dbname="db", password = 'secret',
                   host = 'localhost', port = 5432)

vintageUnitSQL <- "select * from vintage_units"
performanceEventSQL <- "select * from performance_events"
getVintageData(vintageUnitSQL, performanceEventSQL, con = con)

For additional details see help:

help(getVintageData)
help(plotVintageData)
help(printVintageData)
help(aggregateVintageData)
help(createTimeDistanceFunction)

tomasgreif / vintager