Context

This presentation explains the Shiny app you can see in https://viercame.shinyapps.io/week_4_coursera/ The server.R and ui.R code on github can see it in https://github.com/viercame/week4.

Tha data that I use for this analysis correspond to the information of fuel prices in all of the states in Colombia, since jan 2016.

You can get the information from https://www.datos.gov.co/Econom-a-y-Finanzas/Precios-de-Combustibles/7pcy-5vx9

Information about the data

This is the general information of the data.

data <- read.csv("Precios_de_Combustibles.csv", stringsAsFactors = FALSE)

str(data)

## 'data.frame':    150044 obs. of  12 variables:
##  $ CodigoDepartamento: int  5 68 68 68 70 70 73 73 76 5 ...
##  $ NombreDepartamento: chr  "ANTIOQUIA" "SANTANDER" "SANTANDER" "SANTANDER" ...
##  $ CodigoMunicipio   : int  1 1 307 397 708 1 349 449 1 1 ...
##  $ municipio         : chr  "MEDELLIN" "BUCARAMANGA" "GIRON" "LA PAZ" ...
##  $ nombrecomercial   : chr  "ESSO LA 80" "LA GACELA" "ESTACION DE SERVICIO CHIMITA" "ESTACION DE SERVICIO GERSAN" ...
##  $ bandera           : chr  "MOBIL" "MOBIL" "TERPEL" "TERPEL" ...
##  $ direccion         : chr  "Diagonal 80 No.76-95" "Calle 17 No.19-62" "Kilómetro 1 No. 2 - 119 " "CARRERA 3 No. 5-04" ...
##  $ producto          : chr  "BIODIESEL EXTRA" "BIODIESEL EXTRA" "BIODIESEL EXTRA" "BIODIESEL EXTRA" ...
##  $ precio            : int  7390 7221 7040 7650 7528 7450 7380 7390 8500 7420 ...
##  $ estado            : chr  "Activo" "Activo" "Activo" "Activo" ...
##  $ fecharegistro     : chr  "02/18/2016 12:00:00 AM" "02/24/2016 12:00:00 AM" "03/02/2016 12:00:00 AM" "04/02/2016 12:00:00 AM" ...
##  $ periodo           : chr  "02/01/2016 12:00:00 AM" "02/01/2016 12:00:00 AM" "02/01/2016 12:00:00 AM" "02/01/2016 12:00:00 AM" ...

Processing of the data

The data is constricted day per day. For my analysis we calculated the mean, by month + product + state.

library(plyr)

data$fecharegistro <- as.Date(data$fecharegistro,"%m/%d/%Y")
data$periodo <- as.Date(data$periodo,"%m/%d/%Y")

data.summerised <- ddply(data,c("NombreDepartamento","producto","periodo"), summarise, mean=mean(precio))

str(data.summerised)

## 'data.frame':    1957 obs. of  4 variables:
##  $ NombreDepartamento: chr  "AMAZONAS" "AMAZONAS" "AMAZONAS" "AMAZONAS" ...
##  $ producto          : chr  "BIODIESEL EXTRA" "BIODIESEL EXTRA" "BIODIESEL EXTRA" "BIODIESEL EXTRA" ...
##  $ periodo           : Date, format: "2016-01-01" "2016-02-01" ...
##  $ mean              : num  8071 7330 8275 7745 7556 ...

Final product

You can get this kind of plots, using the different options the Shiny app has.

alt text