removed

ons-2005-data-exploration.R

-# Header ###########################################
-# ONS Time Use 2005 data
-#
-# Explorations using R
-#
-# Script for use as part of: The Social Science of Energy Storage (PSY6018)
-#
-# Sheffield/Southampton Centre for Doctoral Training in Energy Storage and its Applications
-# http://www.southampton.ac.uk/engineering/postgraduate/research_degrees/energy_storage_cdt.page
-#
-# Copyright (C) 2014  University of Southampton
-# 
-# Author: Ben Anderson (b.anderson@soton.ac.uk, @dataknut, https://github.com/dataknut) 
-# [Energy & Climate Change, Faculty of Engineering & Environment, University of Southampton]
-# 
-# end header
-
-# To do: -----------------------------------------------------------------
-
-# Prelims -----------------------------------------------------------------
-
-# clear out all old objects etc to avoid confusion
-rm(list = ls()) 
-
-# load required packages
-x <- c("foreign","ggplot2","plyr")
-lapply(x, require, character.only = T)
-
-# if this breaks/fails because they haven't been installed use
-# install.packages("x")
-
-# path to data
-# You will need to have downloaded the data from Dropbox/MOLE
-# You will need to change this!!
-dpath <- "/Users/ben/Documents/Dropbox/energy-storage-dtc/data/"
-
-# Load data -----------------------------------------------------------------
-# First: time use data in long form - this has data in 10 minute time slots
-
-onstu2005_longcsv <- read.csv(paste0(dpath, "/UK-2005-TU-long.csv"))
-
-# Now stop to check what's in it and make sure we understand the format!
-head(onstu2005_longcsv, n = 4)
-
-# check values of main acts (the things people reported doing)
-table("Main acts"= onstu2005_longcsv$pact)
-# check values of location variable (where they reported doing them)
-table("Location"= onstu2005_longcsv$lact)
-# NB: quite a lot of missing
-# check values of months variable (so we see that seasons are represented)
-table("Month"= onstu2005_longcsv$t_month)
-
-# Second: survey data
-onstu2005_surveycsv <- read.csv(paste0(dpath, "/UK-2005-TU-survey-reduced.csv"))
-# check what's in it
-head(onstu2005_surveycsv, n = 2)
-
-
-# Test: Sleep -----------------------------------------------------------------
-onstu2005_longcsv$sleep_p <- 0
-onstu2005_longcsv$sleep_p[onstu2005_longcsv$pact == "sleeping"] <- 1 
-onstu2005_longcsv$sleep_s <- 0
-onstu2005_longcsv$sleep_s[onstu2005_longcsv$sact == "sleeping"] <- 1 
-onstu2005_longcsv$sleep_all <- onstu2005_longcsv$sleep_p + onstu2005_longcsv$sleep_s
-
-sleeping <- ddply(onstu2005_longcsv, c("s_dow", "s_halfhour"), summarise, propn=mean(sleep_all))
-
-ggplot(sleeping, aes(x=s_halfhour, y=propn, colour=s_dow, group=s_dow)) + geom_line()
-
-# Practice: Preparing food -----------------------------------------------------------------
-# Might not require energy of course...
-onstu2005_longcsv$cooking_p <- 0
-onstu2005_longcsv$cooking_p[onstu2005_longcsv$pact == "preparing food" & onstu2005_longcsv$lact != "elsewhere"] <- 1 
-onstu2005_longcsv$cooking_s <- 0
-onstu2005_longcsv$cooking_s[onstu2005_longcsv$sact == "preparing food" & onstu2005_longcsv$lact != "elsewhere"] <- 1 
-onstu2005_longcsv$cooking_all <- onstu2005_longcsv$cooking_p + onstu2005_longcsv$cooking_s
-
-# Need to try to calculate the se somehow if we want error bars?
-cooking_all <- ddply(onstu2005_longcsv, c("s_halfhour"), summarise, 
-                     propn=mean(cooking_all), 
-                     sd=sd(cooking_all),
-                     n=obs(cooking_all))
-
-ggplot(cooking_all, aes(x=s_halfhour, y=propn)) +
-  geom_line(aes(group=1)) +
-  geom_point(size=4) +
-  geom_errorbar(aes(ymin=propn-sd, ymax=propn+sd), width=.2)
-
-
-cooking_dow <- ddply(onstu2005_longcsv, c("s_dow", "s_halfhour"), summarise, propn=mean(cooking_all))
-ggplot(cooking_dow, aes(x=s_halfhour, y=propn, colour=s_dow, group=s_dow)) + geom_line()
-
-# Practice: Car use -----------------------------------------------------------------
-onstu2005_longcsv$car_enj_p <- 0
-onstu2005_longcsv$car_enj_p[onstu2005_longcsv$pact == "travel - car/enjoyment"] <- 1 
-onstu2005_longcsv$car_enj_s <- 0
-onstu2005_longcsv$car_enj_s[onstu2005_longcsv$sact == "travel - car/enjoyment"] <- 1 
-onstu2005_longcsv$car_enj_all <- onstu2005_longcsv$car_enj_p + onstu2005_longcsv$car_enj_s
-
-onstu2005_longcsv$car_oth_p <- 0
-onstu2005_longcsv$car_oth_p[onstu2005_longcsv$pact == "travel - car/other"] <- 1 
-onstu2005_longcsv$car_oth_s <- 0
-onstu2005_longcsv$car_oth_s[onstu2005_longcsv$sact == "travel - car/other"] <- 1 
-onstu2005_longcsv$car_oth_all <- onstu2005_longcsv$car_oth_p + onstu2005_longcsv$car_oth_s                            
-                            
-onstu2005_longcsv$car_shop_p <- 0
-onstu2005_longcsv$car_shop_p[onstu2005_longcsv$pact == "travel - car/shopping"] <- 1 
-onstu2005_longcsv$car_shop_s <- 0
-onstu2005_longcsv$car_shop_s[onstu2005_longcsv$sact == "travel - car/shopping"] <- 1 
-onstu2005_longcsv$car_shop_all <- onstu2005_longcsv$car_shop_p + onstu2005_longcsv$car_shop_s
-
-car_use <- ddply(onstu2005_longcsv, c("s_halfhour"), summarise, shop=mean(car_shop_all), enj=mean(car_enj_all), other=mean(car_oth_all))
-                     
-ggplot(car_use, aes(x=s_halfhour, y=shop)) +
-  aes(group=1) + geom_line()                
-                 
-ggplot(car_use, aes(x=s_halfhour, y=enj)) +
-  aes(group=1) + geom_line() 
-
-# needs to have home as location afterwards for relevance to EV?
-
-
-# merge them
-