Predicting success from music sales data: a statistical and adaptive approach

ABSTRACT

Musical taste is highly individualized and evolves over time in seemingly unpredictable ways. However popular trends emerge as collective and potentially predictable patterns of preference for genre and style. The goal of this paper is to reveal these patterns and extrapolate how a statistically 'average' populace responds to new stimuli.This paper addresses three questions. One is to find statistically meaningful patterns within the data. The next question is if we can predict how long an album will stay in chart, given the first few weeks' sales data, using statistical patterns found from the first question. The last question is to see if a new album's position in chart can be predicted on a certain week in the future (such as the 5th week or 12th week), with the first few weeks' sales data. For this, we used LMS (least mean square) algorithm, a well known adaptive algorithm.This paper uses published bi-weekly sales data from the Billboard magazine, more specifically, the Top Jazz chart. The results show some interesting correlations, one of which emphasizes the role of marketing. According to our findings, it is probably worth a good investment on marketing before starting sales of an album, since the data shows that the higher the starting position of an album is, the longer it is likely to stay in chart.

References

1. Cano, P., Koppenberger, M., and Wack, N., Content-based audio recommendation, In Proceedings of the 13th annual ACM international conference on Multimedia, 2005, 211--212.
2. Berenzweig, A., Logan, B., Ellis, D.P.W., and Whitman, B., A large-scale evaluation of acoustic and subjective music similarity measures, In Proceedings of International Conference on Music Information Retrieval (ISMIR), 2003b, 103--109.
3. Foote, J.T., Content-based retrieval of music and audio, In Proceedings of SPIE, 1997, 138--147.
4. Dhanaraj, R., and Logan, B., Automatic Prediction of Hit Songs, In Proceedings of International Conference on Music Information Retrieval (ISMIR), 2004, 488--491.
5. Montgomery, A.L., and Moe, W.W., Should Record Companies Pay for Radio Airplay? Investigating the Relationship Between Album Sales and Radio Airplay, Working paper, Marketing Dept., The Wharton School, University of Pennsylvania, June 2000.
6. Moe, W.W., and Fader, P.S., Modeling Hedonic Portfolio Products: A Joint Segmentation Analysis of Music Compact Disc Sales, In Journal of Marketing Research, Vol. 38, No. 3, 376--385.
7. Lee, J., Boatwright, P., and Kamakura, W.A., A Bayesian Model for Prelaunch Sales Forecasting of Recorded Music, In Management Science, February 2003, Vol. 49, No. 2, 179--196.
8. Billboard Magazine, Top Jazz chart, from 08/31/2002 to 01/07/2006.
9. Widrow, B., and Stearns, S., Adaptive Signal Processing, Prentice Hall, 1985.