×

Isotonic single-index model for high-dimensional database marketing. (English) Zbl 1429.62689

Summary: While database marketers collect vast amounts of customer transaction data, its utilization to improve marketing decisions presents problems. Marketers seek to extract relevant information from large databases by identifying significant variables and prospective customers. In small databases, they could calibrate logistic regression models via maximum-likelihood methods to determine significant variables and assess customer’s response probability. For large databases, however, this approach becomes computationally too intensive to implement in real-time, and so marketers prefer estimation methods that are scalable to high-dimensional databases. In addition, database marketing is practiced in diverse product-markets, and so marketers prefer probability models that are flexible rather than restrict to specific distributions (e.g., logistic).
To incorporate scalability and flexibility, we propose isotonic single-index models for database marketing. It furnishes the first projective approximation to a general \(p\)-variate function. Its link function is order-preserving (i.e., isotonic), thus encompassing all proper distribution functions. We develop a direct approach for its estimation: we first estimate the orientation of high-dimensional parameter vector without specifying the link function (via sliced inverse regression), and then estimate the non-decreasing link function (via isotonic regression). We illustrate its practical use by analyzing a high-dimensional customer transaction database. This approach yields dimension reduction both column- and row-wise; that is, we not only discover significant variables in a large transaction database, but also prioritize customers into a few distinct groups based on estimated response probability (to enable direct mailing of catalogs).

MSC:

62P20 Applications of statistics to economics
62G07 Density estimation
62J12 Generalized linear models (logistic models)
90B60 Marketing, advertising
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Altham, P. M.E., Improving the precision of estimation by fitting a model, J. Roy. Statist. Soc. Ser. B, 46, 118-119 (1984)
[2] Ayer, M.; Brunk, H. D.; Ewing, G. M.; Reid, W. T.; Silverman, E., An empirical distribution function for sampling with incomplete information, Ann. Math. Statist, 26, 641-647 (1955) · Zbl 0066.38502
[3] Balasubramanian, S.; Gupta, S.; Kamakura, W.; Wedel, M., Modeling large data sets in marketing, Statist. Neerlandica, 52, 303-323 (1998) · Zbl 0941.90044
[4] Barlow, R. E.; Bartholomew, D. J.; Bremner, J. M.; Brunk, H. D., Statistical Inference under Order Restriction (1972), Wiley: Wiley Chichester, UK · Zbl 0246.62038
[5] Berry, M. J.A.; Linoff, G., Data Mining Techniques for Marketing, Sales, and Customer Support (1997), Wiley: Wiley New York, NY
[6] Brillinger, D. R., A generalized linear model with Gaussian regression variables, (Bickel, P. J.; Doksum, K. A.; Hodges, J. L., A Festschrift for Erich L. Lehmann in Honor of His Sixty-Fifth Birthday (1983), Wadsworth: Wadsworth California), 97-114
[7] Bitran, G. R.; Mondschein, S. V., Mailing decisions in the catalog sales industry, Manage. Sci, 42, 1364-1381 (1996) · Zbl 0880.90090
[8] Bult, J. R.; Wansbeek, T., Optimal selection for direct mail, Marketing Sci, 14, 378-394 (1995)
[9] Carroll, R. J.; Fan, J.; Gijbels, I.; Wand, M. P., Generalized partially linear single-index models, J. Amer. Statist. Assoc, 92, 477-489 (1997) · Zbl 0890.62053
[10] Chen, C. H.; Li, K. C., Can SIR be as popular as multiple linear regression?, Statist. Sinica, 8, 289-316 (1998) · Zbl 0897.62069
[11] Cook, R. D.; Lee, H., Dimension reduction in binary response regression, J. Amer. Statist. Assoc, 94, 1187-1200 (1999) · Zbl 1072.62619
[12] Cook, R. D.; Nachtsheim, C. J., Reweighting to achieve elliptically contoured covariates in regression, J. Amer. Statist. Assoc, 89, 592-599 (1994) · Zbl 0799.62078
[13] Cook, R. D.; Weisberg, S., Discussion of sliced inverse regression, J. Amer. Statist. Assoc, 86, 328-332 (1991) · Zbl 1353.62037
[14] Cosslett, S. R., Distribution-free maximum likelihood estimator of the binary choice model, Econometrica, 51, 765-782 (1983) · Zbl 0528.62096
[15] Cox, D. R.; Snell, E. J., Analysis of Binary Data (1989), Chapman & Hall: Chapman & Hall New York, NY · Zbl 0729.62004
[16] Dodson, J. A.; Tybout, A. M.; Sternthal, B., Impact of deals and deal retraction on brand switching, J. Marketing Res, 15, 72-81 (1978)
[17] Duan, N.; Li, K. C., Slicing regressiona link-free regression method, Ann. Statist, 19, 505-530 (1991) · Zbl 0738.62070
[18] Ehrenberg, A. S.C., Laws in marketing, (Bogart, L., Current Controversy in Marketing Research (1969), Markham: Markham Chicago, IL), 141-152
[19] Gifi, A., Nonlinear Multivariate Analysis (1990), Wiley: Wiley New York, NY · Zbl 0697.62048
[20] Gonul, F.; Shi, M. Z., Optimal mailing of catalogsnew methodology using estimable structural dynamic programming models, Manage. Sci, 44, 1249-1262 (1998) · Zbl 0989.90081
[21] Gonul, F.; Kim, B. D.; Shi, M. Z., Mailing smarter to catalog customers, J. Interactive Marketing, 14, 2-16 (2000)
[22] Hall, P.; Huang, L. S., Nonparametric kernel regression subject to monotonicity constraints, Ann. Statist, 29, 624-647 (2001) · Zbl 1012.62030
[23] Horowitz, J. L., Semiparametric Methods in Econometrics (1998), Springer: Springer New York, NY · Zbl 0897.62128
[24] Hsing, T.; Carroll, R. J., An asymptotic theory for sliced inverse regression, Ann. Statist, 20, 1040-1061 (1992) · Zbl 0821.62019
[25] Hughes, A. M., The Complete Database Marketer (1996), McGraw-Hill: McGraw-Hill New York, NY
[26] Ichimura, H., Semiparametric least squares (SLS) and weighted SLS estimation of single-index models, J. Econom, 58, 71-120 (1993) · Zbl 0816.62079
[27] Kiefer, J.; Wofowitz, J., Consistency of the maximum likelihood estimator in the presence of infinitely many incidental parameters, Ann. Math. Statist, 27, 887-906 (1956) · Zbl 0073.14701
[28] Leeflang, P.; Wittink, D. R.; Wedel, M.; Naert, P. A., Building Models for Marketing Decisions (2000), Kluwer Academic Publishers: Kluwer Academic Publishers Boston, MA
[29] Li, K. C., Sliced inverse regression for dimension reduction, J. Amer. Statist. Assoc, 86, 316-342 (1991), (with discussions) · Zbl 0742.62044
[30] Li, K.C., 2000. High Dimensional Data Analysis Via the SIR/PHD Approach. Unpublished manuscript dated April 6, 2000 obtained at the Internet site www.stat.ucla.edu kcli/sir-PHD.pdf; Li, K.C., 2000. High Dimensional Data Analysis Via the SIR/PHD Approach. Unpublished manuscript dated April 6, 2000 obtained at the Internet site www.stat.ucla.edu kcli/sir-PHD.pdf
[31] McLachlan, G.; Peel, D., Finite Mixture Models (2000), Wiley: Wiley New York, NY · Zbl 0963.62061
[32] Naik, P. A.; Raman, K., Understanding the impact of synergy in multimedia communications, J. Marketing Res, 40, 4, 375-388 (2003)
[33] Naik, P. A.; Tsai, C. L., Partial least squares estimator for single-index models, J. Roy. Statist. Soc. Ser. B, 62, 763-771 (2000) · Zbl 0957.62060
[34] Naik, P. A.; Tsai, C. L., Single-index model selections, Biometrika, 88, 3, 821-832 (2001) · Zbl 0988.62042
[35] Powell, J. L.; Stock, J. H.; Stoker, T. M., Semiparametric estimation of index coefficients, Econometrica, 57, 1403-1430 (1989) · Zbl 0683.62070
[36] Schmittlein, D. C.; Morrison, D. G.; Colombo, R., Counting your customerswho are they and what will they do next?, Manage. Sci, 33, 1-24 (1987)
[37] Sharma, S., Applied Multivariate Techniques (1996), Wiley: Wiley New York
[38] Silverman, B. W., Density Estimation (1986), Chapman & Hall: Chapman & Hall London · Zbl 0617.62042
[39] Simonoff, J. S., Smoothing Methods in Statistics (1996), Springer: Springer New York, NY · Zbl 0859.62035
[40] Statistical Fact Book, 2000. The Direct Marketing Association, 22nd Edition. New York, NY.; Statistical Fact Book, 2000. The Direct Marketing Association, 22nd Edition. New York, NY.
[41] Stoker, T. M., Consistent estimation of scaled coefficients, Econometrika, 54, 1461-1481 (1986) · Zbl 0628.62105
[42] Wald, A., Note on the consistency of the maximum likelihood estimate, Ann. Math. Statist, 20, 595-601 (1949) · Zbl 0034.22902
[43] Wedel, M.; Kamakura, W., Market Segmentation: Conceptual and Methodological Foundations (2000), Kluwer Academic: Kluwer Academic Boston, MA
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.