VARIABLES PERSONALES QUE INTERVIENEN EN EL CONFORT TÉRMICO

2.3. CONFORT TÉRMICO

2.3.2. VARIABLES PERSONALES QUE INTERVIENEN EN EL CONFORT TÉRMICO

The success of direct marketing campaigns relies on the accuracy of models used to predict promotional response. Due to limited promotional budgets and an effort to reduce waste caused by unredeemed promotions, marketers strive to target only those households with a high probability of promotional response. This study examines the relationship between entire purchase path (spending and visits path) of households before receiving a mass or customized promotion and the household’s total expenditures at the retailer during the promotional period.

We utilize twenty weeks of household store-level spending and visits data before exposure to a direct retailer promotion and use key features from this purchase path as predictor variables of a household’s spending during the promotional period. The contribution of this study can be

summarized as follows: (1) we propose a flexible, semi-parametric functional data approach to identify purchase path features from a household’s spending history, (2) we compare the RFM and

purchase path feature and explain extra information obtained from purchase paths, (3) we provide new managerial insights on purchase path that significantly explain a household’s total spending

during customized and mass promotions, (4) we provide a parsimonious model that adapts to randomness in the presence of sparse and noisy spending information, (4) we develop a method to help marketing managers determine the total spending of targeted households well in advance of (at least four weeks prior to) the start of a direct promotion.

The results of this study suggest that essential information in a household’s purchase path

provides information that is more useful for predicting spending during a promotional period than RFM, household demographics and loyalty card promotional variables. As such, the characteristics of a household’s spending path can explain why a given household responds more favorably to a

we find that households with increasing trend in their spending and visits before the promotion spend more during both types of promotions. This is perhaps because increasing spending and store visits over time indicates satisfaction and convenience with shopping at the store. The aggregated information in the monetary spending (M) and frequency of visits (F) could not capture the nuances of spending and visits variations over time that can be captured using entire purchase path information.

We also find that patterns in household’s store visit moderates the relationships between other campaign and store related variables. For instance, two households with similar prior campaign exposure or similar number favorite coupons in the campaign the one with increasing visits over time would spend significantly higher during the direct promotion. Our analysis is robust to potential selection bias as we control for this bias using IPTW weights and use Weighted Least Squared (WLS) regression to estimate the parameters of interest. We indeed found that the assignment of households to the treatment groups to be biased and un weighted models lead to biased estimates of the model coefficients. The predictive accuracy of sales during the promotions is also significantly higher for models incorporating RFM and purchase path features than the models with only RFM and other demographic information. For all predictive models under consideration, the models that account for the weighted models perform significantly better than unweighted models.

We also conducted a simulation experiment to understand the performance of the proposed FDA framework in a controlled environment with varying randomness in the households’

consumption rate. We find that the representation of the spending path curves (determined by a roughness penalty) fitted on the household data positively improves out-of-sample predictions. Therefore, the data-driven approach adapts to the variations in the sparse household data at the

disposal of the managers for building marketing response models. The results of the simulation show that the performance of the functional approach remains consistently better than the model with accumulated RFM data.

Although this paper provides many insights for managers that utilize direct promotions, the study is not without limitations. First, we were unable to control for household purchases in other distribution channels that might affect spending at the focal retailer during the promotional period. While less than 1% of total grocery sales in the US occurred online during the period of analysis (2005-07), future research must integrate purchasing data from multiple channels to provide a more comprehensive understanding of a household’s spending path.

Second, although we adjusted for potential selection bias by using a propensity score matching procedure, a randomized controlled experiment would be helpful in complementing our empirical investigation.

Third, we selected 20 weeks prior spending history of the households, and in some case, they were targeted with multiple promotions during the response period. Although, we control for multiple promotions by using it as a control in our explanatory model the future research must develop methods to disentangle and estimate the impact of simultaneous promotional efforts on a household.

Finally, our findings are context specific in that we investigate household spending at a grocery retailer. However, the two-stage functional data approach proposed in this study is exploratory and does not a priori assume patterns in the data that impact the response variable. Hence, it can easily be generalized to other predictive scenarios involving longitudinal data and can adapt to variations in that data to give the best predictive outcomes. Given these strengths, the approach has been applied to the context of online auctions (Jank and Shmueli 2006), virtual stock

markets (Foutz and Jank 2010), and product diffusion (Sood et al. 2009). However, the breadth of its applications is endless, and future studies could apply this approach to other types of longitudinal data and similar relationships, such as the relationship between browsing history and response to online advertisements or brand switching, to name a few.

References

Abdi, Herve, and Lynne J. Williams (2010), “Principal component analysis,” WIREs Comp Stat, 2, 433–459.

Ailawadi, Kusum L., Bari A. Harlam, Jacques César, and David Trounce (2006). “Promotion

Profitability for a Retailer: The Role of Promotion, Brand, Category, and Store Characteristics,” Journal of Marketing Research, 43(4), 518-535.

Ali, Özden G., Serpil Sayın, Tom van Woensel, and Jan Fransoo (2009). “SKU demand forecasting in the presence of promotions,” Expert Systems with Applications, 36(10), 12340-12348. Allenby, Greg M., Robert P. Leone, and Lichung Jen (1999). “A Dynamic Model of Purchase

Timing with Application to Direct Marketing,” Journal of the American Statistical

Association, 94(446), 365-374.

Anderson, Eric T., Gavan J. Fitzsimons, and Duncan Simester (2006). “Measuring and Mitigating the Costs of Stockouts,” Management Science, 52(11), 1751-1763.

Baesens, Bart, Stijn Viaene, Dirk Van den Poel, Jan Vanthienen, and Guido Dedene (2002). “Bayesian neural network learning for repeat purchase modelling in direct marketing,” European Journal of Operational Research, 138(1), 191-211.

Bell, David R., Daniel Corsten, and George Knox (2011). “From Point of Purchase to Path to Purchase: How Preshopping Factors Drive Unplanned Buying,” Journal of

Marketing, 75(1), 31-45.

Berger, Paul, and Thomas L. Magliozzi (1992). “The effect of sample size and proportion of buyers

in the sample on the performance of list segmentation equations generated by regression analysis,” Journal of Direct Marking, 6(1), 13-22.

Blattberg, Robert C., Gary D. Eppen and Joshua Lieberman (1981). “A Theoretical and Empirical Evaluation of Price Deals for Consumer Nondurables,” Journal of Marketing, 45(1), 116. Breiter, Andreas, and Arold Huchzermeier (2014). “Promotion Planning and Supply Chain

Contracting in a High-Low Pricing Environment,” Production and Operations Management, 24(2), 219-236.

Buckinx, Wouter, and Dirk Van den Poel (2005). “Customer base analysis: partial defection of

behaviourally loyal clients in a non-contractual FMCG retail setting,” European Journal of Operational Research, 164(1), 252-268.

Cooper, Lee G., Penny Baron, Wayne Levy, Michael Swisher, and Paris Gogos (1999). “PromoCast™: A New Forecasting Method for Promotion Planning,” Marketing

Science, 18(3), 301-316.

Cui, Geng, Man L. Wong, and Hon-Kwong Lui (2006). “Machine Learning for Direct Marketing Response Models: Bayesian Networks with Evolutionary Programming,” Management

Science, 52(4), 597-612.

Damon, Amy L., Robert P. King, and Ephraim Leibtag (2013). “First of the month effect: Does it apply across food retail channels?” Food Policy, 41, 18-27.

Diepen, Merel Van, Bas Donkers and Philip H. Franses (2009). “Dynamic and Competitive Effects of Direct Mailings: A Charitable Giving Application,” Journal of Marketing

Research, 46(1), 120-133.

Divakar, Suresh, Brian T. Ratchford, and Venkatesh Shankar (2005). “Practice Prize Article—

CHAN4CAST: A Multichannel, Multiregion Sales Forecasting Model and Decision Support System for Consumer Packaged Goods,” Marketing Science, 24(3), 334-350.

Donkers, Bas, Richard R. Paap, Jedid-Jah Jonker, and Philip Hans Franses (2006). “Deriving target selection rules from endogenously selected samples,” Journal of Applied

Economics, 21(5), 549-562.

Forbes Insights and Turn (2015). “The Rise of the New Marketing Organization,” (accessed March

18 2017), [available at https://images.forbes.com/forbesinsights/StudyPDFs/Turn- The_Rise_of_the_New_Marketing_Organization-REPORT.pdf].

Foutz, Natasha, and Wolfgang Jank (2010). “Research Note—Prerelease Demand Forecasting for Motion Pictures Using Functional Shape Analysis of Virtual Stock Markets,” Marketing

Science, 29(3), 568-579.

Heilman, Carrie M., Kent Nakamoto, and Ambar G. Rao (2002). “Pleasant Surprises: Consumer

Response to Unexpected In-Store Coupons,” Journal of Marketing Research, 39(2), 242- 252.

Huang, Tao, Robert Fildes, and Didier Soopramanien (2014). “The value of competitive

information in forecasting FMCG retail product sales and the variable selection problem,” European Journal of Operations Research, 237(2), 738-748.

Imai, Kosuke (2017). “MatchIt: Nonparametric Preprocessing for Parametric Causal Inference,” (accessed May 9, 2017), [available at http://gking.harvard.edu/matchit].

Jank, Wolfgang, and Galit Shmueli (2006). “Functional Data Analysis in Electronic Commerce Research,” Statistical Science, 21(2), 155-166.

Jing, Xiaoqing, and Michael Lewis (2011). “Stockouts in Online Retailing,” Journal of Marketing

Research, 48(2), 342-354.

Johnson, Joseph M., Gerard J. Tellis, and Edward H. Ip (2013). “To Whom, When, and How Much to Discount? A Constrained Optimization of Customized Temporal Discounts,” Journal of

Retailing, 89(4), 361-373.

Joffe, M., Ten Have, T., Feldman, H. and Kimmel, S. (2004). Model Selection, Confounder Control, and Marginal Structural Models. The American Statistician, 58(4), pp.272-279. Kahn, Barbara E., and David Shmittlein (1992). “The relationship between purchases made on

promotion and shopping trip behavior,” Journal of Retailing, 68(3), 294-315.

Keele, Luke J. (2010). "Rbounds: An R Package For Sensitivity Analysis with Matched Data.", available at http://www.personal.psu.edu/ljk20/rbounds.html

Lee, Wing Yee, Paul Goodwin, Robert Fildes, Kostantinos Nikolopoulos, and Michael Lawrence (2007). “Providing support for the use of analogies in demand forecasting tasks,” International Journal of Forecasting, 23(3), 377-390.

Leeflang, Peter S. H., Harald J. van Heerde, and Dick R. Wittink (2002). “How Promotions Work:

SCAN*PRO-Based Evolutionary Model Building,” Schmalenbach Business Review, 54 (Summer), 198-220.

Markets and Markets (2014). “Retail Analytics Market worth USD 5.1 Billion USD by 2020” (accessed March 18 2017), [available at

McCaffrey, D., Griffin, B., Almirall, D., Slaughter, M., Ramchand, R. and Burgette, L. (2013). A tutorial on propensity score estimation for multiple treatments using generalized boosted models. Statistics in Medicine, 32(19), pp.3388-3414.

Mithas, Sunil, and Krishnan, M.S. (2009). “From Association to Causation via a Potential Outcomes Approach,” Information Systems Research, 20(2), 295-313.

Nikolopoulos, Konstantinos (2010). “Forecasting with quantitative methods: the impact of special events in time series,” Applied Economics, 42(8), 947-955.

Ramsay, James, and Bernard W. Silverman (2002). Functional data analysis, (2nd ed.) New York: Springer.

Rao, Vithala R., and Joel H. Steckel (1995). “Selecting, evaluating, and updating prospects in direct mail marketing,” Journal of Direct, Data and Digital Marketing Practice, 9(2), 20-

31.

Rosenbaum, Paul R. and Donald B. Rubin (1983). “Assessing Sensitivity to a Binary Covariate in an Observational Study with Binary Outcome”, Journal of the Royal Statistical Society.

Series B, 45(2), 212-218.

Rosenbaum, Paul R. and Donald B. Rubin (1984). “Reducing Bias in Observational Studies Using

Subclassification on the Propensity Score,” Journal of the American Statistical Association, 79(387), 516-524.

Schweidel, David A., and George Knox (2013). “Incorporating direct marketing activity into latent attrition models,” Marketing Science, 32(3), 471-487.

Sood, Ashish, Gareth M. James, and Gerard J. Tellis (2009). “Functional Regression: A New Model for Predicting Market Penetration of New Products,” Marketing Science, 28(1), 36-

51.

Stilley, Karen M., Jeffrey J. Inman, and Kirk L. Wakefield (2010). “Spending on the Fly: Mental Budgets, Promotions, and Spending Behavior,” Journal of Marketing, 74(3), 34-47.

Trapero, Juan R., Nikolaos Kourentzes, and Robert Fildes (2014). "On The Identification Of Sales Forecasting Models In The Presence Of Promotions," Journal of Operational Research Society, 66(2), 299-307.

Ullah, Shahid, and Caroline F. Finch (2013). “Applications of functional data analysis: A systematic review,” BMC Medical Research Methodology, 13(1), 43.

VanderWeele, Tyler (2015), Explanation in causal inference (1st ed.). Oxford University Press. Venkatesan, Raj., and Paul W. Farris (2012). “Measuring and Managing Returns from Retailer-

Customized Coupon Campaigns,” Journal of Marketing, 76(1), 76-94.

Wang, Jane-Ling, Jeng-Min Chiou, and Hans-Georg Muller (2016). “Review of functional data analysis,” Annual Review of Statistics and Its Application, 3, 257-295.

Xavier, Dreze, Patricia Nisol, and Naufel J. Vilcassim (2004). “Do Promotions Increase Store

Expenditures? A Descriptive Study of Household Shopping Behavior,” Quantitative Marketing and Economics, 2(1), 59-92.

Xiong, Guiyang and Sundar Bharadwaj (2014). “Prerelease Buzz Evolution Patterns and New Product Performance,” Marketing Science, 33(3), 401-421.

Zahavi, Jacob, and Nissan Levin (1997). “Applying neural computing to target marketing,” Journal of Direct Marketing, 11(1), 5-22.

CHAPTER 3: MODELING PHYSICIANS’ PRESCRIPTION CHOICES

In document El uso de los espacios abiertos frente a las condiciones climáticas. El caso de dos parques residenciales de Monterrey (página 52-59)