Who Watches What: Forecasting Viewership for the Top 100 TV Networks

  • Denis KhryashchevEmail author
  • Alexandru Papiu
  • Jiamin Xuan
  • Olivia Dinica
  • Kyle Hubert
  • Huy Vo
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11917)


TV advertising makes up more than one third of total ad spending, and is transacted based on forecast ratings and viewership. Over time, forecast accuracy has decreased due to fragmentation of consumer behavior. Through a comprehensive study we find that an assortment of models combined with an ensemble method leads to better accuracy than any single method. This results in an 11% improvement over a naive baseline method, across 100 of the largest networks.


TV networks Time series Forecasting XGboost Fourier Facebooks Prophet Seasonal averaging 



This work was supported in part by Pitney Bowes 3100041700, Alfred P. Sloan Foundation G-2018-11069, and NSF award 1827505.


  1. 1.
    Oster, E.: TVs share of ad spend expected to continue its decline this year. (2018)
  2. 2.
    Arvidsson, J.: Forecasting on-demand video viewership ratings using neural networks (2014)Google Scholar
  3. 3.
    Napoli, P.M.: The unpredictable audience: an exploratory analysis of forecasting error for new prime-time network television programs. J. Advert. 30(2), 53–60 (2001)MathSciNetCrossRefGoogle Scholar
  4. 4.
    Cooley, J.W., Tukey, J.W.: An algorithm for the machine calculation of complex fourier series. Math. Comput. 19(90), 297–301 (1965)MathSciNetCrossRefGoogle Scholar
  5. 5.
    Oppenheim, A., Schafer, R., Buck, J.: Discrete-Time Signal Processing. Prentice Hall, Upper Saddle River (1999)Google Scholar
  6. 6.
    Hubert, K.: The hidden story behind TV’s ratings decline. (2017)
  7. 7.
    Hindman, D.B., Wiegand, K.: The big three’s prime time decline: a technological and social context. J. Broadcast. Electron. Media 52(1), 119–135 (2008)CrossRefGoogle Scholar
  8. 8.
    Hunter III, S.D., Chinta, R., Smith, S., Shamim, A., Bawazir, A.: Moneyball for TV: a model for forecasting the audience of new dramatic television series. Stud. Media Commun. 4(2), 13–22 (2016)Google Scholar
  9. 9.
    Zigmond, D., et al.: When viewers control the schedule: measuring the impact of digital video recording on TV viewership. In: Key Issues Forums at ARF Audience Measurement Conference (2009)Google Scholar
  10. 10.
    Weber, R.: Methods to forecast television viewing patterns for target audiences. In: Communication Research in Europe and Abroad Challenges of the First Decade. De-Gruyter, Berlin (2002)Google Scholar
  11. 11.
    Neagu, R.: Forecasting television viewership: a case study. GE Global Research, 2003GRC039 (2003)Google Scholar
  12. 12.
    Pagano, R., Quadrana, M., Cremonesi, P., Bittanti, S., Formwentin, S., Mosconi, A.: Prediction of TV ratings with dynamic models. In: ACM Workshop on Recommendation Systems for Television and Online Video, RecSysTV (2015)Google Scholar
  13. 13.
    Meyer, D., Hyndman, R.J.: The accuracy of television network rating forecasts: the effects of data aggregation and alternative models. Model Assist. Stat. Appl. 1(3), 147–155 (2005)MathSciNetzbMATHGoogle Scholar
  14. 14.
    Nikolopoulos, K., Goodwin, P., Patelis, A., Assimakopoulos, V.: Forecasting with cue information: a comparison of multiple regression with alternative forecasting approaches. Eur. J. Oper. Res. 180(1), 354–368 (2007)CrossRefGoogle Scholar
  15. 15.
    Wang, C., Goossens, D., Vandebroek, M.: The impact of the soccer schedule on TV viewership and stadium attendance: evidence from the Belgian Pro League. J. Sports Econ. 19(1), 82–112 (2018)CrossRefGoogle Scholar
  16. 16.
    Gambaro, M., Larcinese, V., Puglisi, R., Snyder Jr., J.M.: Is soft news a turn-off? Evidence from Italian TV news viewership (2017)Google Scholar
  17. 17.
    Belo, R., Ferreira, P., de Matos, M., Reis, F.: The impact of time-shift TV on TV viewership and on ad consumption: results from both natural and randomized experiments. A theory of the economics of time. Econ. J. 81(324), 828–846 (2016)Google Scholar
  18. 18.
    Taylor, S.J., Letham, B.: Forecasting at scale. Am. Stat. 72(1), 37–45 (2018)MathSciNetCrossRefGoogle Scholar
  19. 19.
    Huber, P.J.: Robust statistics. In: Lovric, M. (ed.) International Encyclopedia of Statistical Science. Springer, Heidelberg (2011). Scholar
  20. 20.
    Chen, T., Guestrin, C.: XGBoost: a scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 785–794. ACM (2016)Google Scholar
  21. 21.
    Shen, W., Babushkin, V., Aung, Z., Woon, W.L.: An ensemble model for day-ahead electricity demand time series forecasting. In: Proceedings of the Fourth International Conference on Future Energy Systems, pp. 51–62. ACM (2013)Google Scholar
  22. 22.
    Taylor, J.W., McSharry, P.E., Buizza, R., et al.: Wind power density forecasting using ensemble predictions and time series models. IEEE Trans. Energy Convers. 24(3), 775 (2009)CrossRefGoogle Scholar
  23. 23.
    Kourentzes, N., Barrow, D.K., Crone, S.F.: Neural network ensemble operators for time series forecasting. Expert Syst. Appl. 41(9), 4235–4244 (2014)CrossRefGoogle Scholar
  24. 24.
    Makridakis, S., Hibon, M.: The M3-competition: results, conclusions and implications. Int. J. Forecast. 16(4), 451–476 (2000)CrossRefGoogle Scholar
  25. 25.
    Makridakis, S., Spiliotis, E., Assimakopoulos, V.: The M4 competition: results, findings, conclusion and way forward. Int. J. Forecast. 34(4), 802–808 (2018)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Denis Khryashchev
    • 1
    Email author
  • Alexandru Papiu
    • 2
  • Jiamin Xuan
    • 2
  • Olivia Dinica
    • 2
  • Kyle Hubert
    • 2
  • Huy Vo
    • 3
  1. 1.The Graduate CenterCity University of New YorkNew YorkUSA
  2. 2.SimulmediaNew YorkUSA
  3. 3.The City College of New YorkCity University of New YorkNew YorkUSA

Personalised recommendations