Searching for the picture: forecasting UK cinema admissions using Google Trends data

Abstract

This article investigates whether Google Trends search information can improve forecasts of cinema admissions. Using monthly data for the United Kingdom for the period 1 January 2004 to 31 December 2008, we examine various forecasting models that incorporate Google Trends search information. We find clear evidence that Google Trends data on searches relevant to cinema visits (as opposed to searches for specific films) do have the potential to increase the accuracy of cinema admissions forecasting models.

Keywords:

• cinema admissions
• Internet search engine
• forecasting

JEL Classification:

• C22
• L82

Notes

¹ Termed ‘present casting’ by Choi and Varian (2009).

² The data source is the UK Film Council Statistical Yearbooks, various issues, available online at http://www.ukfilmcouncil.org.uk/.

³ Obtainable via http://www.google.co.uk/trends.

⁴ The results cover the period 4 January 2004 to 25 April 2010 and were obtained on 28 April 2010. Full details on how the series are produced can be found at http://www.google.co.uk/intl/en/trends/about.html.

⁵ We also tried using company names such as Odeon, Cineworld and Vue.

⁶ All estimations and testings were undertaken using the PcGive module of the Oxmetrics 4 software suite (Timberlake Consultants, London, UK). Full results for these and other search terms are available from the authors on request.