1st Data Analysis Contest Using R

Sponsored by Nestoria

Nestoria is an online property search engine used by more than 2.5 million Internet visitors a month. Nestoria manages a large database of millions of eligible property ads for sale and rent in 8 countries. Property ads are updated by the most relevant property portals on a daily or weekly basis. The ads, or listings, are presented online in sites by location and property type, as in this example.
Objective: find value in Nestoria's data using R Nestoria and Lokku Labs are the proud sponsors of the 1st Data Analysis Contest Using R as part of the 3rd Conference of R Users in Madrid on November 17th and 18th of 2011. Nestoria and Lokku Labs aim to improve the understanding of the public of the value of its databases. The company aims to engage a few brilliant statisticians in the expectation that their work promotes the value of their house prices data.
1,500€ in prizes... and more

The Winner will receive a cheque for 1,000€ and the Second Winner, a cheque for 500€, both in addition of a certificate of the granted award. The Judges might declare the prizes vacant and void if the quality or public interest is deemed insufficient by the sole and exclusive decision of the Judges.

Additionally, one or more of the contestants might be invited to a job interview in London, UK for a 3 months internship in the offices of the company. All contestants will be sent a certificate of their submission to the Contest with the title of their work.

From 18/07 to 15/10The contest begins on July 18th 2011. It ends on October 15th 2011.
Open to all R users
  • The Contest is open to anyone with a good command of R and Statistics from any country, for any number of entries per private individual and/or institution.
  • All or a material part of the creative process or design will be carried out and replicable entirely on R language and software.
  • If all or part of the entry has already been published, you will attach the explicit written permission to submit by the editor/s of the original publication of your work along with an explicit written authorisation to publish the logo or brand of the publication in the Media Activity described below.
  • The works of the entry will include R code, basic instructions to operate it, a sample of the data used in the work, graphics and visual material, text describing the approach, methodology and results of the statistical analysis and a summary and live presentation explaining the interest of the work for the scientific community and the general public.
Participation & further infoIn order to enter the competition, send solutions, or other inquiries, send an email to Participants are encouraged to register for statistical purpuses using the following form.
Data downloadData can be downloaded from this link. The file readme.txt provides additional low level information about the data.
Data description

House prices are the driving forces behind some of the major economic successes and crisis world-wide in the last few years. Nestoria aggregate house prices data of Australia, Brazil, France, Germany, India, Italy, Spain and the UK. The Sponsor hopes to inspire capable Statisticians to help our visitors understand the phenomena and trends of house prices across geographies over time.

Currently for each area or location Nestoria compute all the properties on offer in that area, calculate the geometric mean, remove all properties beyond two median absolute deviations from the geometric mean, and calculate a new geometric mean. The result is plotted on graphs available on pages like this one. Computations are currently carried out in Perl language. The Sponsor feels that the of house prices and their visual representation could be improved significantly with the aid of scientific rigour and visual creativity for visitor to their web sites on average unfamiliar to house prices and statistics.

The Sponsor would like to present the data of house prices to casual visitors of its web sites in an original, innovative and self-evident or visual (infographic) way trends, features or correlations of house prices unknown to the average casual visitor

In order to encourage the creativity and rigour of statisticians, the organizers of the 3rd Conference of Users of R and Nestoria provide

  • Prices of offer of sale and rent residential properties in 8 countries of the international footprint of Nestoria from the last 3 to 12 months
  • Script in R to query the Nestoria API for individual prices
  • Raw data provided by Nestoria

Analysis of Nestoria data in combination with data of third party like disposable income, transactional house prices or any other statistics provided by government or NGO is explicitly welcome but not necessary.

Media activityBy participating in this Contest, you agree and hereby grant the Organizers and the Sponsor permission to use, copy, modify, distribute, and publicly display your Contest submissions for any purpose, such as, but not limited to, press and media communications, without any compensation or attribution to you. You also agree to participate in any media or promotional activity regarding the Contest.
  • Rubén Martínez Sánchez (Nestoria), chairman
  • Miguel Ángel Rodríguez Muíños (Servizo Galego de Saúde)
  • Virgilio Gómez Rubio (Universidad de Castilla-La Mancha)
  • Juan José Gibaja Martíns (Universidad de Deusto)
  • Emilio Torres Manzanera (Universidad de Oviedo)
Judging and Selection of Winners

Entries will be judged by the based on the following judging criteria:

  1. scientific merit
  2. creativity and ingenuity
  3. apparent simplicity and meaningfulness for casual visitors of the material presented
  4. scalability of the code in R to large databases and web sites in production with millions of visitors per month
  5. whether the submission adheres to these Rules

Authors of the entries shortlisted according to the criteria above will be invited to present live their entries publicly to the Judges during the 3rd Conference of Users of R in Madrid on November 17th and 18th of 2011. The live presentation is a requirement of the Contest to be awarded the Prizes.

  • By entering the Contest, you agree to comply with and be bound by the Rules.
  • Failure to comply with these Rules may result in your disqualification from the Contest.
  • You further agree to comply with and be bound by the decisions of the Judges, which are final and binding in all respects.
  • You expressly waive any and all rights to recourse to other procedures to litigate against the Sponsor or the Organizers.
  • In case of interpretation conflict between the English and the Spanish versions of these rules, the English version will be used.
Intellectual PropertyBy registering for the Contest and/or submitting an entry you agree explicitly that all components of your scientific work is original and that it will be published under the GNU General Public License is a free, copyleft license for software and other kinds of works on the web sites of Nestoria, Lokku Labs and the Organizers of the 3rd Conference of Users of R in Madrid on November 17th and 18th of 2011 for free without any further liability.

© Copyright 2011. All rights reserved. Nestoria is the trademark of Lokku Ltd, a company registered in England. Lokku Labs is a web site featuring a few projects about online property search.