Abstract
The state space model is a very general model, mostly used to specify structural time-series models. Structural time-series models explicitly specify trends and seasonality along with other relevant influences. Under the classical Box-Jenkins time-series approach, in contrast, trends and seasonal influences are removed before estimating the core model. At the heart of state space models is the specification of one or more unobserved time series α, the states, to describe observed time series y. States can serve several purposes in state space models. For example, one may use states to capture unobserved trends in the observed time series y, to model unobserved (latent) variables such as goodwill, or to specify unobserved time-varying response parameters. Given the increasing availability of time-series data in marketing (Pauwels et al. 2004 and Chaps. 3 and 4), the importance of hard-to-measure variables such as goodwill (e.g., Naik et al. 1998), and the knowledge that response parameters may change dramatically over time (e.g., Osinga et al. 2010), state space models are highly relevant for marketing.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
See Vol. I, Sect. 8.2.4.
References
Aravindakshan, A., Peters, K., Naik, P.A.: Spatiotemporal allocation of advertising budgets. J.Mark. Res. 49, 1–14 (2012)
Bruce, N.I., Peters, K., Naik, P.A.: Discovering how advertising grows sales and builds brands. J.Mark. Res. 49, 793–806 (2012)
Du, R.Y., Kamakura, W.A.: Quantitative Trendspotting. J. Mark. Res. 49, 514–536 (2012)
Durbin, J., Koopman, S.J.: Time Series Analysis by State Space Methods. Oxford University Press, Oxford (2001)
Jap, S.D., Naik, P.A.: Bidanalyzer: a method for estimation and selection of dynamic bidding models. Mark. Sci. 27, 949–960 (2008)
Kolsarici, C., Vakratsas, D.: Category-versus brand-level advertising messages in a highly regulated environment. J. Mark. Res. 47, 1078–1089 (2010)
Koopman, S.J.: Exact initial Kalman filtering and smoothing for nonstationary time series models. J. Am. Stat. Assoc. 92, 1630–1638 (1997)
Liu, Y., Shankar, V.: The dynamic impact of product-harm crises on brand preference and advertising effectiveness: an empirical analysis of the automobile industry. Manag. Sci. 61, 2514–2535 (2015)
McQuarie, A., Tsai, C.: Regression and Time Series Model Selection. World Scientific, Singapore (1998)
Naik, P.A.: Marketing dynamics: a primer on estimation and control. Foundations and Trends® in Marketing. 9, 175–266 (2015)
Naik, P.A., Mantrala, M.K., Sawyer, A.: Planning pulsing media schedules in the presence of dynamic advertising quality. Mark. Sci. 17, 214–235 (1998)
Naik, P.A., Raman, K.: Understanding the impact of media synergy in multimedia communications. J. Mark. Res. 40, 375–388 (2003)
Naik, P.A., Shi, P., Tsai, C.: Extending the Akaike information criterion to mixture regression models. J. Am. Stat. Assoc. 102, 244–254 (2007)
Nerlove, M., Arrow, K.: Optimal advertising policy under dynamic conditions. Economica. 29, 129–142 (1962)
Osinga, E.C., Leeflang, P.S.H., Srinivasan, S., Wieringa, J.E.: Why do firms invest in consumer advertising with limited sales response? A shareholder perspective. J. Mark. 75(1), 109–124 (2011)
Osinga, E.C., Leeflang, P.S.H., Wieringa, J.E.: Early marketing matters: a time-varying parameter approach to persistence modeling. J. Mark. Res. 47, 173–185 (2010)
Pauwels, K.H., Currim, I., Dekimpe, M.G., Hanssens, D.M., Mizik, N., Ghysels, E., Naik, P.A.: Modeling marketing dynamics by time series econometrics. Market. Lett. 15, 167–183 (2004)
Van Heerde, H.J., Mela, C.F., Manchanda, P.: The dynamic effect of innovation on market structure. J. Mark. Res. 41, 166–183 (2004)
West, M., Harrison, J.: Bayesian Forecasting and Dynamic Models. Springer, New York (1997)
Xie, J.X., Song, M., Sirbu, M., Wang, Q.: Kalman filter estimation of new product diffusion models. J. Mark. Res. 34, 378–393 (1997)
Acknowledgements
The author gratefully acknowledges the support of the Netherlands Organisation for Scientific Research (NWO) under grant number 016.135.234. Also, the author thanks Prasad Naik for commenting on an earlier version of this chapter and for sharing his extensive set of materials on state space models.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this chapter
Cite this chapter
Osinga, E.C. (2017). State Space Models. In: Leeflang, P., Wieringa, J., Bijmolt, T., Pauwels, K. (eds) Advanced Methods for Modeling Markets. International Series in Quantitative Marketing. Springer, Cham. https://doi.org/10.1007/978-3-319-53469-5_5
Download citation
DOI: https://doi.org/10.1007/978-3-319-53469-5_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-53467-1
Online ISBN: 978-3-319-53469-5
eBook Packages: Business and ManagementBusiness and Management (R0)