Abstract Advertising is commonly used as a major marketing tool by many firms to support their new product release. Quantifying the effectiveness of pre-release advertising campaigns, however, is both challenging, since no sales data are available, and costly, because of the need to conduct consumer surveys. This is especially true for movie industry for which the majority of advertising dollars are spent before the movie’s release. Using the recent availability of online data on consumer search behaviors on a popular website dedicated to the movie industry, we construct a consumer interest measure to help the decision makers evaluate their advertising effectiveness. We build a dynamic model to show how this cost-efficient measure of consumer interest can be used to capture pre-release advertising dynamics, and the impact of advertising on weekly movie revenues starting with the opening week. Using a procedure based on Naik, Mantrala, and Sawyer (1998) and controlling for endogeneity, we estimate response, forgetting and wear-out of pre-release and post-release advertising for a large sample of movies. Our empirical results show that advertising generates goodwill, and that the resulting goodwill is associated with increased revenues. Our results suggest that big-budget movies can increase advertising effectiveness by making small adjustments to their advertising dynamics, but that spending more advertising dollars will likely not be effective. The data on the evolution of consumer interest can be a valuable and inexpensive tool for measuring advertising effectiveness and understanding sales dynamics.
{"title":"Role of Advertising and Consumer Interest in the Motion Picture Industry","authors":"Dongling Huang, Andrei Strijnev, B. Ratchford","doi":"10.1515/roms-2014-0005","DOIUrl":"https://doi.org/10.1515/roms-2014-0005","url":null,"abstract":"Abstract Advertising is commonly used as a major marketing tool by many firms to support their new product release. Quantifying the effectiveness of pre-release advertising campaigns, however, is both challenging, since no sales data are available, and costly, because of the need to conduct consumer surveys. This is especially true for movie industry for which the majority of advertising dollars are spent before the movie’s release. Using the recent availability of online data on consumer search behaviors on a popular website dedicated to the movie industry, we construct a consumer interest measure to help the decision makers evaluate their advertising effectiveness. We build a dynamic model to show how this cost-efficient measure of consumer interest can be used to capture pre-release advertising dynamics, and the impact of advertising on weekly movie revenues starting with the opening week. Using a procedure based on Naik, Mantrala, and Sawyer (1998) and controlling for endogeneity, we estimate response, forgetting and wear-out of pre-release and post-release advertising for a large sample of movies. Our empirical results show that advertising generates goodwill, and that the resulting goodwill is associated with increased revenues. Our results suggest that big-budget movies can increase advertising effectiveness by making small adjustments to their advertising dynamics, but that spending more advertising dollars will likely not be effective. The data on the evolution of consumer interest can be a valuable and inexpensive tool for measuring advertising effectiveness and understanding sales dynamics.","PeriodicalId":35829,"journal":{"name":"Review of Marketing Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2015-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1515/roms-2014-0005","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66992969","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A standard result in the literature on durable goods is that if buyers have foresight regarding a durable goods seller’s future pricing strategies, then the seller cannot perfectly price discriminate intertemporally. In a finite time horizon framework where buyers are large and finite in number, Bagnoli, Salant, and Swierzbinski (1989) constructed numerical examples in which the seller of the durable product is able to perfectly price discriminate across its consumers. This paper extends the examples in Bagnoli et al. to a general model of a durablegoods seller selling over a finite horizon to any finite number of individually significant consumers with different reservation prices for the product and provides a complete characterization of the equilibrium pricing strategies of the seller.
{"title":"When Can a Durable Goods Seller Price Discriminate Intertemporally?","authors":"Sreya Kolay","doi":"10.1515/roms-2014-0007","DOIUrl":"https://doi.org/10.1515/roms-2014-0007","url":null,"abstract":"A standard result in the literature on durable goods is that if buyers have foresight regarding a durable goods seller’s future pricing strategies, then the seller cannot perfectly price discriminate intertemporally. In a finite time horizon framework where buyers are large and finite in number, Bagnoli, Salant, and Swierzbinski (1989) constructed numerical examples in which the seller of the durable product is able to perfectly price discriminate across its consumers. This paper extends the examples in Bagnoli et al. to a general model of a durablegoods seller selling over a finite horizon to any finite number of individually significant consumers with different reservation prices for the product and provides a complete characterization of the equilibrium pricing strategies of the seller.","PeriodicalId":35829,"journal":{"name":"Review of Marketing Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2015-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1515/roms-2014-0007","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66993031","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract Prior research has shown that consumer online search can serve as an indicator of consumers’ general interest. However, it has not yet differentiated between consumers’ category- and product-level interest. This article argues that online search volume for terms submitted to search engines can indicate consumer interest in a general product category or a specific product. A consumer may search for a product category (e.g. “Broadway shows”) to obtain information about several products within the category before making a purchase decision. A consumer may also search for a specific product (e.g. “Shrek the Musical”) to obtain more information about that product. The objective of this article is to investigate the relationship between aggregate volumes of category- vs. product-level search terms and ticket sales of Broadway shows. Methodologically, we employ a fixed-effects regression model and specify weekly sales of each show as a function of its product- and category-level search term volume. We also account for potential effect changes after a critical industry-wide event, the Tony Awards ceremony. Our results show that product-level search is an important driver before and after the Tony Awards, yet category-level search is important only after the Tony Awards. Our findings suggest important implications for marketing communications. Managers can use these results to manage or generate consumer category- and product-level interest more effectively at various points of time.
{"title":"Examining the Dynamics of Consumer Interest and Live Performance Event TicketSales in the Presence of a Critical Industry-Wide Event","authors":"P. H. Tseng, Gauri Kulkarni","doi":"10.1515/roms-2012-0002","DOIUrl":"https://doi.org/10.1515/roms-2012-0002","url":null,"abstract":"Abstract Prior research has shown that consumer online search can serve as an indicator of consumers’ general interest. However, it has not yet differentiated between consumers’ category- and product-level interest. This article argues that online search volume for terms submitted to search engines can indicate consumer interest in a general product category or a specific product. A consumer may search for a product category (e.g. “Broadway shows”) to obtain information about several products within the category before making a purchase decision. A consumer may also search for a specific product (e.g. “Shrek the Musical”) to obtain more information about that product. The objective of this article is to investigate the relationship between aggregate volumes of category- vs. product-level search terms and ticket sales of Broadway shows. Methodologically, we employ a fixed-effects regression model and specify weekly sales of each show as a function of its product- and category-level search term volume. We also account for potential effect changes after a critical industry-wide event, the Tony Awards ceremony. Our results show that product-level search is an important driver before and after the Tony Awards, yet category-level search is important only after the Tony Awards. Our findings suggest important implications for marketing communications. Managers can use these results to manage or generate consumer category- and product-level interest more effectively at various points of time.","PeriodicalId":35829,"journal":{"name":"Review of Marketing Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1515/roms-2012-0002","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66992838","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mohammad Arzaghi, E. Berndt, James C. Davis, A. Silk
Abstract We address a puzzle surrounding the shift from bundling to unbundling of U.S. advertising agency services and the slow pace of change over several decades. We model an agency’s decision as a tradeoff between the fixed cost to the advertiser of establishing a relationship with an agency and pecuniary economies of scale from media services provision. Using micro-data from the U.S. Census of Services for 1982-2007, we find agencies are more likely to unbundle with increasing size, diversification and higher media prices, and less likely with increasing age, larger media volume, and an interaction between media prices and volume.
{"title":"The Unbundling of Advertising Agency Services: An Economic Analysis","authors":"Mohammad Arzaghi, E. Berndt, James C. Davis, A. Silk","doi":"10.1515/1546-5616.1164","DOIUrl":"https://doi.org/10.1515/1546-5616.1164","url":null,"abstract":"Abstract We address a puzzle surrounding the shift from bundling to unbundling of U.S. advertising agency services and the slow pace of change over several decades. We model an agency’s decision as a tradeoff between the fixed cost to the advertiser of establishing a relationship with an agency and pecuniary economies of scale from media services provision. Using micro-data from the U.S. Census of Services for 1982-2007, we find agencies are more likely to unbundle with increasing size, diversification and higher media prices, and less likely with increasing age, larger media volume, and an interaction between media prices and volume.","PeriodicalId":35829,"journal":{"name":"Review of Marketing Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2012-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1515/1546-5616.1164","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66785814","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract The existing theoretical literature on mass customization maintains that customization reduces product differentiation and intensifies price competition. In contrast, operations management studies argue that customization serves primarily to differentiate a company from its competitors. Interactive involvement of the customer in product design creates an affective relationship with the firm, relaxing price competition. This paper provides a model that incorporates consumer involvement to explain the phenomena described in the operations management literature.Two firms on the Hotelling line compete for a continuum of consumers with heterogeneous brand preferences. An exogenously given fraction of consumers is potentially interested in customization. Consumer benefits from customization are the rewards from a special shopping experience and the value of product customization (a better fitting product); these benefits are higher for consumers located closer to the customizing brand. When a consumer purchases a customized product, he/she incurs waiting costs. Each firm simultaneously decides whether to offer standard products, customized products, or both, and then engage in price competition. I show that customization increases product differentiation, leading to less intense price competition. Depending on the parameter values, in equilibrium either both firms offer customized products, one firm offers customized products and the other standard and customized products, or one firm offers customized products and the other standard products. I perform comparative statics analysis with respect to the fraction of consumers interested in customization, the waiting costs, and the fixed cost of customization.
{"title":"Competitive Effects of Mass Customization","authors":"O. Loginova","doi":"10.1515/1546-5616.1139","DOIUrl":"https://doi.org/10.1515/1546-5616.1139","url":null,"abstract":"Abstract The existing theoretical literature on mass customization maintains that customization reduces product differentiation and intensifies price competition. In contrast, operations management studies argue that customization serves primarily to differentiate a company from its competitors. Interactive involvement of the customer in product design creates an affective relationship with the firm, relaxing price competition. This paper provides a model that incorporates consumer involvement to explain the phenomena described in the operations management literature.Two firms on the Hotelling line compete for a continuum of consumers with heterogeneous brand preferences. An exogenously given fraction of consumers is potentially interested in customization. Consumer benefits from customization are the rewards from a special shopping experience and the value of product customization (a better fitting product); these benefits are higher for consumers located closer to the customizing brand. When a consumer purchases a customized product, he/she incurs waiting costs. Each firm simultaneously decides whether to offer standard products, customized products, or both, and then engage in price competition. I show that customization increases product differentiation, leading to less intense price competition. Depending on the parameter values, in equilibrium either both firms offer customized products, one firm offers customized products and the other standard and customized products, or one firm offers customized products and the other standard products. I perform comparative statics analysis with respect to the fraction of consumers interested in customization, the waiting costs, and the fixed cost of customization.","PeriodicalId":35829,"journal":{"name":"Review of Marketing Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2012-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1515/1546-5616.1139","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66785347","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract Although there is enormous evidence that reference levels influence preferences, conjoint models, one of the most successful marketing research tools, assume that preferences depend on the absolute levels of attributes. In this paper we investigate the relevance of reference effects in two settings, compositional or self-explicated models in experimental studies 1 and 2, and decompositional or choice-based models in experimental study 3. In particular, we introduce a simple modification of the traditional self-explicated conjoint model which permits dependence of preference on reference levels. By eliciting gains and losses from expectations the model is adaptable to changes in respondents' reference points, which the traditional model is incapable of. Reference options are found to clearly affect subject choices in studies 1 and 2. In addition, the reference dependent self-explicated model is found to offer useful predictions when reference points are manipulated in study 1, and improve on predictions of its traditional counterpart when reference points are measured in study 2. In contrast, in study 3, the choice-based model’s diagnostics and predictions are found to be robust to reference point manipulations. Taken together, these results suggest that the self-explicated model is more suited than the choice-based model to understanding and predicting how respondents make judgments relative to reference points because reference points and gains and losses from reference levels are more salient in the self-explicated model. We discuss implications for managers constructing conjoint models in product-market settings wherein reference points are changing due to new product introductions or marketing efforts.
{"title":"Reference Dependence and Conjoint Analysis","authors":"Brennan Davis, Imran S. Currim, R. Sarin","doi":"10.1515/1546-5616.1141","DOIUrl":"https://doi.org/10.1515/1546-5616.1141","url":null,"abstract":"Abstract Although there is enormous evidence that reference levels influence preferences, conjoint models, one of the most successful marketing research tools, assume that preferences depend on the absolute levels of attributes. In this paper we investigate the relevance of reference effects in two settings, compositional or self-explicated models in experimental studies 1 and 2, and decompositional or choice-based models in experimental study 3. In particular, we introduce a simple modification of the traditional self-explicated conjoint model which permits dependence of preference on reference levels. By eliciting gains and losses from expectations the model is adaptable to changes in respondents' reference points, which the traditional model is incapable of. Reference options are found to clearly affect subject choices in studies 1 and 2. In addition, the reference dependent self-explicated model is found to offer useful predictions when reference points are manipulated in study 1, and improve on predictions of its traditional counterpart when reference points are measured in study 2. In contrast, in study 3, the choice-based model’s diagnostics and predictions are found to be robust to reference point manipulations. Taken together, these results suggest that the self-explicated model is more suited than the choice-based model to understanding and predicting how respondents make judgments relative to reference points because reference points and gains and losses from reference levels are more salient in the self-explicated model. We discuss implications for managers constructing conjoint models in product-market settings wherein reference points are changing due to new product introductions or marketing efforts.","PeriodicalId":35829,"journal":{"name":"Review of Marketing Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2012-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1515/1546-5616.1141","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66786220","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Though individual demand and supply equations can readily be expressed in logit models, closed-form solutions for equilibrium shares and prices are intractable due to the presence of products of polynomial and exponential terms. This hinders the employment of logit models in theoretical studies, and also makes it difficult to develop reduced-form expressions for share and price as a function of exogenous variables for use in empirical studies. In this paper we propose that a mathematical function called the LambertW be employed in solving logit models for equilibrium shares and prices. We derive closed form solutions for price and share in both the monopoly case as well as in the presence of competition. In the competitive case, the prices of the focal firm and the competitor are dependent on each other; hence the equilibrium prices are endogenous and need to be determined simultaneously. To solve this issue, we provide a simple technique that researchers can employ to derive the optimal prices for both the focal firm and the competitor simultaneously.
{"title":"Solving Share Equations in Logit Models Using the LambertW Function","authors":"A. Aravindakshan, B. Ratchford","doi":"10.2202/1546-5616.1120","DOIUrl":"https://doi.org/10.2202/1546-5616.1120","url":null,"abstract":"Though individual demand and supply equations can readily be expressed in logit models, closed-form solutions for equilibrium shares and prices are intractable due to the presence of products of polynomial and exponential terms. This hinders the employment of logit models in theoretical studies, and also makes it difficult to develop reduced-form expressions for share and price as a function of exogenous variables for use in empirical studies. In this paper we propose that a mathematical function called the LambertW be employed in solving logit models for equilibrium shares and prices. We derive closed form solutions for price and share in both the monopoly case as well as in the presence of competition. In the competitive case, the prices of the focal firm and the competitor are dependent on each other; hence the equilibrium prices are endogenous and need to be determined simultaneously. To solve this issue, we provide a simple technique that researchers can employ to derive the optimal prices for both the focal firm and the competitor simultaneously.","PeriodicalId":35829,"journal":{"name":"Review of Marketing Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2010-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.2202/1546-5616.1120","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"68628467","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The main purpose of generic advertising is to enlarge the total market demand rather than capturing further slices from competitors. Several studies point out emergence of free-riding behavior under independent contribution and suggest use of coordination mechanism. However, existing literature does not shed light on the conditions under which generic advertising can be detrimental (beneficial) to stronger firms weakening (strengthening) their competitive advantage. Also, under a setting including both price and brand advertising competition, coordination in generic advertising has not been unraveled. In order to deal with such issues, we consider a one-stage duopoly game in which two firms jointly spend in generic advertising and, simultaneously, compete by setting price and brand advertising. Under independent contribution, we show that when generic advertising effectiveness is high or differences between firms are overall small, weaker firms free-riding lowers the profit difference between the two firms and, sometimes, leads the stronger firm to make even lower profit. On the other hand, in presence of low generic advertising effectiveness or high asymmetries between firms, the stronger firm takes more advantage from generic advertising and increases the profit gap. Under coordination we consider two commonly used mechanisms to share generic advertising expenditure, i.e. fixed percentage (FP) mechanism and sales-proportional (SP) mechanism. We discover that neither mechanism is always dominant in terms of industry profit. In fact, under symmetry, SP mechanism always drives firms to spend more in generic advertising. However, this will result in a higher industry profit if either price or brand advertising competition is fierce or generic advertising effectiveness is low. On the other hand, FP mechanism can guarantee a bigger pie when both price and brand advertising competition is mild and generic advertising is highly effective. Numerical analysis under asymmetry seems to confirm our findings.
{"title":"Generic Advertising, Brand Advertising and Price Competition: An Analysis of Free-Riding Effects and Coordination Mechanisms","authors":"P. Roma, G. Perrone","doi":"10.2202/1546-5616.1122","DOIUrl":"https://doi.org/10.2202/1546-5616.1122","url":null,"abstract":"The main purpose of generic advertising is to enlarge the total market demand rather than capturing further slices from competitors. Several studies point out emergence of free-riding behavior under independent contribution and suggest use of coordination mechanism. However, existing literature does not shed light on the conditions under which generic advertising can be detrimental (beneficial) to stronger firms weakening (strengthening) their competitive advantage. Also, under a setting including both price and brand advertising competition, coordination in generic advertising has not been unraveled. In order to deal with such issues, we consider a one-stage duopoly game in which two firms jointly spend in generic advertising and, simultaneously, compete by setting price and brand advertising. Under independent contribution, we show that when generic advertising effectiveness is high or differences between firms are overall small, weaker firms free-riding lowers the profit difference between the two firms and, sometimes, leads the stronger firm to make even lower profit. On the other hand, in presence of low generic advertising effectiveness or high asymmetries between firms, the stronger firm takes more advantage from generic advertising and increases the profit gap. Under coordination we consider two commonly used mechanisms to share generic advertising expenditure, i.e. fixed percentage (FP) mechanism and sales-proportional (SP) mechanism. We discover that neither mechanism is always dominant in terms of industry profit. In fact, under symmetry, SP mechanism always drives firms to spend more in generic advertising. However, this will result in a higher industry profit if either price or brand advertising competition is fierce or generic advertising effectiveness is low. On the other hand, FP mechanism can guarantee a bigger pie when both price and brand advertising competition is mild and generic advertising is highly effective. Numerical analysis under asymmetry seems to confirm our findings.","PeriodicalId":35829,"journal":{"name":"Review of Marketing Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2010-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.2202/1546-5616.1122","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"68628560","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract The logit model is the most popular tool in estimating demand for differentiated products. In this model, the outside good plays a crucial role because it allows consumers to stop buying the differentiated good altogether if all brands simultaneously become less attractive (e.g. if a simultaneous price increase occurs). But practitioners lack data on the outside good when only aggregate data are available. The currently accepted procedure is to assume a “market potential” that implicitly defines the size of the outside good (i.e. the number of consumers who considered the product but did not purchase); in practice, this means that an endogenous quantity is approximated by a reasonable guess thereby introducing the possibility of an additional source of error and, most importantly, bias. We provide two contributions in this paper. First, we show that structural parameters can be substantially biased when the assumed market potential does not approximate the outside option correctly. Second, we show how to use panel data techniques to produce unbiased structural estimates by treating the market potential as an unobservable in both the simple and the random coefficients logit demand model. We explore three possible solutions: (a) controlling for the unobservable with market fixed effects, (b) specifying the unobservable to be a linear function of product characteristics, and (c) using a “demeaned regression” approach. Solution (a) is feasible (and preferable) when the number of goods is large relative to the number of markets, whereas (b) and (c) are attractive when the number of markets is too large (as in most applications in Marketing). Importantly, we find that all three solutions are nearly as effective in removing the bias. We demonstrate our two contributions in the simple and random coefficients versions of the logit model via Monte Carlo experiments and with data from the automobile and breakfast cereals markets.
{"title":"Eliminating the Outside Good Bias in Logit Models of Demand with Aggregate Data","authors":"Dongling Huang, C. Rojas","doi":"10.1515/roms-2013-0016","DOIUrl":"https://doi.org/10.1515/roms-2013-0016","url":null,"abstract":"Abstract The logit model is the most popular tool in estimating demand for differentiated products. In this model, the outside good plays a crucial role because it allows consumers to stop buying the differentiated good altogether if all brands simultaneously become less attractive (e.g. if a simultaneous price increase occurs). But practitioners lack data on the outside good when only aggregate data are available. The currently accepted procedure is to assume a “market potential” that implicitly defines the size of the outside good (i.e. the number of consumers who considered the product but did not purchase); in practice, this means that an endogenous quantity is approximated by a reasonable guess thereby introducing the possibility of an additional source of error and, most importantly, bias. We provide two contributions in this paper. First, we show that structural parameters can be substantially biased when the assumed market potential does not approximate the outside option correctly. Second, we show how to use panel data techniques to produce unbiased structural estimates by treating the market potential as an unobservable in both the simple and the random coefficients logit demand model. We explore three possible solutions: (a) controlling for the unobservable with market fixed effects, (b) specifying the unobservable to be a linear function of product characteristics, and (c) using a “demeaned regression” approach. Solution (a) is feasible (and preferable) when the number of goods is large relative to the number of markets, whereas (b) and (c) are attractive when the number of markets is too large (as in most applications in Marketing). Importantly, we find that all three solutions are nearly as effective in removing the bias. We demonstrate our two contributions in the simple and random coefficients versions of the logit model via Monte Carlo experiments and with data from the automobile and breakfast cereals markets.","PeriodicalId":35829,"journal":{"name":"Review of Marketing Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2010-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1515/roms-2013-0016","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66992886","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Rick L. Andrews, M. Brusco, Imran S. Currim, Brennan Davis
This study compares the effectiveness of statistical model-based (MB) clustering methods with that of more commonly used non model-based (NMB) procedures in three important contexts: the traditional cluster analysis problem in which a set of consumer characteristic variables is used to form segments; clusterwise regression, in which response parameters from a regression form the basis of segments, and bicriterion clustering problems, which arise when managers wish to form market segments jointly on the basis of a set of characteristics and response parameters from a regression. If the managers primary objective is to forecast responses for segments of holdout consumers for whom only characteristics are available, NMB procedures perform better than MB procedures. However, if it is important to understand the true segmentation structure in a market as well as the nature of the regression relationships within segments, the MB procedure is clearly preferred. Bicriterion segmentation methods are shown to be advantageous when there is at least some concordance between segments derived from different bases. Insights from the simulation study shed new light on a social marketing application in the area of segmenting and profiling overweight youths.
{"title":"An Empirical Comparison of Methods for Clustering Problems: Are There Benefits from Having a Statistical Model?","authors":"Rick L. Andrews, M. Brusco, Imran S. Currim, Brennan Davis","doi":"10.2202/1546-5616.1117","DOIUrl":"https://doi.org/10.2202/1546-5616.1117","url":null,"abstract":"This study compares the effectiveness of statistical model-based (MB) clustering methods with that of more commonly used non model-based (NMB) procedures in three important contexts: the traditional cluster analysis problem in which a set of consumer characteristic variables is used to form segments; clusterwise regression, in which response parameters from a regression form the basis of segments, and bicriterion clustering problems, which arise when managers wish to form market segments jointly on the basis of a set of characteristics and response parameters from a regression. If the managers primary objective is to forecast responses for segments of holdout consumers for whom only characteristics are available, NMB procedures perform better than MB procedures. However, if it is important to understand the true segmentation structure in a market as well as the nature of the regression relationships within segments, the MB procedure is clearly preferred. Bicriterion segmentation methods are shown to be advantageous when there is at least some concordance between segments derived from different bases. Insights from the simulation study shed new light on a social marketing application in the area of segmenting and profiling overweight youths.","PeriodicalId":35829,"journal":{"name":"Review of Marketing Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2010-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.2202/1546-5616.1117","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"68628265","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：