This is day 12 of "One CSV, 30 stories":http://blog.whatfettle.com/2014/10/13/one-csv-thirty-stories/ a series of articles exploring "price paid data":https://www.gov.uk/government/statistical-data-sets/price-paid-data-downloads from the Land Registry found on GOV.UK. The code for this and the other articles is available as open source from "GitHub":https://github.com/psd/price-paid-data
"Yesterday":http://blog.whatfettle.com/2014/11/02/one-csv-thirty-stories-11-calendar/ I picked up the "d3":http://d3js.org/ library and today is another hack based on one of Mike Bostock's examples, this time for "stacked and grouped bar charts":http://bl.ocks.org/mbostock/3943967. The example uses randomly generated data, so I needed to tweak the code to read in price-paid data.
Thus far we've been working with tab-separated files, but d3 works well with comma-separated files where the first row is titles. The "pricebands.awk":https://github.com/psd/price-paid-data/blob/master/bin/pricebands.awk script counts prices, dividing the count of transactions per-year into bands of interesting prices from "Day 8":http://blog.whatfettle.com/2014/10/25/one-csv-thirty-stories-8-heatmap-meh/. Here's a snippet of that code:
print "Year,£60k and under,£61k to £250k,£251k to £500k,£501k to £1M,Over £1M"
...
sub("-.*$", "", date);
price = $2 / 1000;
if (price <= 60) { counts[0]++ }
else if (price <= 250) { counts[1]++ }
else if (price <= 500) { counts[2]++ }
else if (price <= 1000) { counts[3]++ }
else { counts[4]++ }
When the script is run against our price-paid data gives the following CSV:
bc. Year,£60k and under,£61k to £250k,£251k to £500k,£501k to £1M,Over £1M
1995,452525,303592,8538,1269,175
1996,518596,397198,12608,1796,300
1997,533210,505534,19528,2909,529
1998,469642,529691,23943,3475,696
1999,460087,671743,38262,5786,1138
2000,379997,677118,47535,8087,1812
2001,342353,811221,64924,10455,2228
2002,275118,943248,100839,15521,2958
2003,173954,934596,118623,16924,2838
2004,97122,977390,159680,23260,3996
2005,46969,832235,146136,23289,3846
2006,33781,1027533,211643,36286,6355
2007,22906,949966,236803,43975,8564
2008,15427,488654,112597,22462,5038
2009,18378,471070,103899,21411,4636
2010,18550,465895,134915,31332,7194
2011,21655,469188,127024,30472,7264
2012,20127,461865,133592,31313,7456
2013,23776,545992,170000,41920,10668
2014,14324,347393,116810,30444,7976
I'm new to d3, so it did take me a little while to work out how to add axes, and then re-scale them when the histogram transitioned from stacked bars to groups of bars:
I'm quite pleased with the result: they clearly show the loss of the sub-£60k price-band and the fall in transactions from 2008.
There's still quite a lot to say about the how the transactions and prices are distributed over time, but I think "tomorrow":http://blog.whatfettle.com/2014/11/06/one-csv-thirty-stories-13-postcodes/ we need to start looking into where all these houses are actually located.