In this issue, Cui et al. (2018) show how the quantity and quality of user‐generated Facebook data can be used to enhance product forecasts. The intent of this note is to show how another type of user‐generated content—customer search data, specifically one obtained from Google Trends—can be used to reduce out‐of‐sample forecast errors. Based on our work with an online retailer, we bolster Cui et al. (2018) result by showing that adding customer search data to time series models improves out‐of‐sample forecast errors.
BooneT.GaneshanR.HicksR. L.. 2015. Incorporating Google Trends Data into Sales Forecasting. Foresight: Int. J. Appl. Forecast, Summer (38): 9–14.
2.
CastleJ. L.NicholasN. W. P.HendryD. F.. 2009. Nowcasting is not just contemporaneous forecasting. Natl Inst. Econ. Rev.210(1): 71–89.
3.
ChoiH., VarianH.. 2009. Predicting the present with Google Trends. Technical Report, Google Inc. Available at http://google.com/googleblogs/pdfs/google_predicting_the_present.pdf (accessed date May 1, 2017).
4.
ChoiH.VarianH.. 2012. Predicting the present with Google Trends. Econ. Rec.88(s1):2–9.
5.
CuiR.GallinoS.MorenoA.ZhangD.. 2018. The operational value of social media information. Prod. Oper. Manage.27(10): 1749–1769.
6.
DimpflT.JankS.. 2016. Can internet search queries help to predict stock market volatility?Eur. Financ. Manag.22(2): 171–192.
LimniosC.YouH.. 2016. Can Google Trends actually improve housing market forecasts? SSRN Working paper. Available at https://ssrn.com/abstract=2886705 or https://doi.org/10.2139/ssrn.2886705 (accessed date May 1, 2017).
