Market segmentation by maximum likelihood clustering using choice elasticities

We determine market segments by clustering households on the basis of their average choice elasticities across purchases and brands w.r.t. price, sales promotion and brand loyalty. The cluster analysis technique used is a maximum likelihood method which allows varying size and orientation and assumes constant volume. Elasticities originate from choice models with alternatively linear and nonlinear utility functions. Choice models are estimated on the basis of household scanner data. Segments are interpreted by means of multiple discriminant analysis and multinomial logit models whose predictors are elasticities of predictors and external variables (i.e. number of purchases, number of brands bought, income and household size), respectively.     

Please or Squeeze? Brand performance implications of constrained and unconstrained multi-item promotions

Existing research on price promotions typically investigates the effects of single-product promotions. However, brand managers often promote multiple items simultaneously to bolster the performance of a complete brand line. This paper investigates the brand performance implications of such multi-item promotions. We distinguish between two types of multi-item promotions: traditional, unconstrained multi-item promotions that grant a discount irrespective of the size or composition of the purchase set, and constrained multi-item promotions that require consumers to purchase specific quantities (self-bundling) or even predetermined combinations (fixed bundling) of promoted items. Using a normative consumer decision model, the authors analyze the relative effectiveness of these promotions in terms of unit sales and revenue impact and show that promotion performance is moderated by consumers’ preference for variety. The authors find that unconstrained multi-item promotions outperform single-item promotions, especially in markets without variety-seeking. In markets with variety seeking, unconstrained multi-item promotions are not as effective in stealing extra sales from competition because consumers prefer to spread their purchases across all items, including competitive ones. However, in those markets, self-bundling promotions are particularly effective because they force consumers to adopt larger amounts of the promoted items at the expense of competitors. Finally, in neither type of market does fixed bundling substantially improve promotion performance.     

Brand positioning under lexicographic choice rules

This paper discusses a brand positioning model in which two brands of a product are to be positioned in a price-quality space under a new behavioral assumption. This assumption asserts that customers determine the highest-quality product within their reservation price and purchase it, provided its quality does not fall short of a minimum standard. The model also includes producers' costs that are incurred for delivering a certain quality. We first delineate reaction functions for the optimal location of one brand, give a location of its competitor. We then show that Nash equilibria do not exist as long as price and quality are both variable. Finally, we consider a two phase model: in the first phase, the duopolists sequentially choose their quality levels under the assumption that both competitors know that in the second phase, a Nash equilibrium in prices follows. Single-variable mathematical programming formulations are presented to solve the problem. A numerical example is also given to illustrate the working of the model.     

A new stochastic ultrametric tree unfolding methodology for assessing competitive market structure and deriving market segments

We present a new methodology for simultaneously assessing competitive market structure and deriving market segments. A hierarchical or ultrametric tree representation is estimated in a maximum likelihood framework from collected paired-comparison choice data. The derived tree portrays both brands and consumers/households/segments as terminal nodes, where the ‘closer’ a brand is to a particular consumer/household/segment in the tree, the higher the predicted probability of that consumer/household/segment choosing that particular brand. This paper initially presents an introduction to the problem of market structure assessment. We review the extensive marketing literature on market structure and survey several competing methodologies. The proposed stochastic ultrametric tree unfolding methodology is technically described and several program options are indicated. An illustration of the proposed methodology is presented with respect to paired comparison choice data collected from a convenience sample involving the over-the-counter analgesics market. Finally, several areas for future research are identified.     

Measuring the effects of new brand introduction on inter-brand strategic interaction

This paper empirically investigates the effects of a new brand introduction on the market structure of a product market, i.e., the nature of competition among existing brands. The model is based on the discrete choice multinomial logit model in which the household-specific brand intercepts are decomposed into brand locations in attribute space and households' importance weights for these attributes. The formulation also incorporates the effects of marketing variables on brand choice behavior. The relative proximity of brands in the attribute space provides a measure of the intensity of competitive rivalry among brands. A new brand introduction results in an additional location in this space. Three consequences of the introduction are investigated. (i) Impact on brand locations of extant brands. (ii) Change in the importance weights assigned by households to the attributes. (iii) Effects on households' sensitivities to marketing activities. As some or all of the above could change, these three effects result in eight combinations that need to be studied. The general model is one in which all three components are allowed to change and this nests the remaining seven specifications. The specification that is most consistent with the data can, therefore, be isolated via a statistical test of nested hypotheses. Identifying the consequences of the introduction for market structure has implications for managerial action. An empirical application of the model to household scanner panel data using liquid laundry detergents is provided.     

Household risk aversion and portfolio choices

In practice, stock investment is one of the most important decisions made by households. The primary goal of this paper is to explain family investment decisions under the assumptions of household member’s preferences and efficient risk sharing based on the collective household model. In particular, by examining the absolute (relative) risk aversion of the household welfare function, we demonstrate how household’s portfolio allocation in stocks changes with family wealth. We examine two types of preference heterogeneity between family members: parameter heterogeneity and functional form heterogeneity. This study offers an alternative explanation of household portfolio choice corresponding with the observation that wealthier households tend to hold greater share of their wealth in risky assets. Specifically, if two decision-makers have standard constant relative risk aversion preference with different relative risk aversions in a household, family’s relative risk aversion decreases as household wealth increases (decreasing relative risk aversion).     

Estimating household purchase rates for consumer nondurable goods

Understanding the purchase rates of households for frequently purchased packaged goods is an important element in developing effective marketing strategies. Previous researchers have attempted to estimate these rates by assuming that the time between purchases is a random variable that follows some common parametric probability distribution such as the exponential or Weibull distribution. Recent research has shown that for many frequently purchased packaged goods, the interpurchase times cannot be adequately described by these commonly used probability distributions. In this study we demonstrate how household purchase rates can be estimated in a robust manner using a generalized semiparametric approach that obviates the need for specifying a parametric form for the distribution of interpurchase times. The motivation being that often there is no theory of household purchase behaviour that specifies a priori the probability distribution underlying the interpurchase times. Our empirical results indicate that, for the data analysed, the household purchase rates exhibit a regular pattern that cannot be recovered by probability distributions often used in previous research. Further, marketing actions taken by sellers do indeed influence household purchase behaviour.     

不同价格模式下考虑需求波动性和差异化服务的网约车平台定价决策

针对单个平台两种品牌网约车的最优定价问题,考虑平台服务质量的差异化和市场需求波动性,分别建立动态价格、差异化价格和静态价格模式下的网约车动态服务模型,运用多元函数和泛函的条件极值求得两种品牌网约车的最优定价策略。研究发现,平台最优动态价格和差异化价格均随需求波动时长单调变化,而最优静态价格并非单调。此外,平台提高差异化服务时,两种品牌网约车的最优价格均提高,但高服务质量的网约车会有更高的提价幅度;固定佣金报酬率增大时,平台最优价格均提高,但边际损失成本较大的网约车会有更高的提价幅度。最后,通过数值仿真对不同价格模式下的平台利润进行比较和灵敏度分析,并发现平台利润在市场需求稳定时差异不大。     

Inventory Control with a Randomly Available Discount Purchase Price

A mathematical model is developed for a basic single-commodity storage system in which the stock controller is faced with a randomly fluctuating purchase price. High- and low-price periods are defined and a purchasing policy is specified which depends upon both the number of units in storage and the price period in which the system is operating. For the case of random demand the limiting stochastic behaviour of the system is obtained and the cost of operating the system is derived.     

品牌的非对称需求价格弹性模型:以中国手机行业为例

本文提出了计算品牌的需求价格弹性的非对称市场反应函数模型,从而使对向上的价格弹性和向下的价格弹性的分别考量成为可能。根据此模型,本文对中国手机行业进行了实证研究,共计算了GSM市场的13个品牌和CDMA市场的9个品牌的向上和向下的价格弹性,并且通过方差分析对国内外手机品牌的价格弹性进行了比较。结果发现,在GSM和CDMA市场上国内外手机品牌的价格弹性都存在显著差异,反映了国内外手机品牌的市场力量的差异。最后根据研究结果为我国国内手机品牌提出对策建议。     

A rational random behaviour model of choice

In this paper, a multinomial-Dirichlet-geometric model of consumer brand choice is developed. This individual-level stochastic choice model is derived as an extension of Theil's theory of rational random behaviour. These behavioural assumptions permit modelling of changes in likelihood of purchase as consumers are confronted with environmental factors whose occurrence and exact nature could not be anticipated at the planning stage of a shopping trip. Moreover, the model allows for uncertainties about future events which might affect actual choice to be built into the choice process alongside a traditional choice model which reflects preferences and/or utilities (and potential uncertainties surrounding them). Empirical results using consumer diary purchase panel data indicate a strong superiority of the model developed compared with previous models which assume stationary preference vectors.     

Components of optimal price under logit demand

Demand for durables can be modeled using a logit framework in which a customer chooses one brand from several alternatives, or buys nothing at all. In this framework, optimal prices for competing brands can be expressed as a system of non-linear equations, which, however, do not have closed form solutions. Although the optimal price can be determined by numerical search, the solution offers limited understanding of its components. In this article, we develop a linear approximation of the Nash equilibrium optimal price of a brand as its marginal cost plus a weighted sum of: (1) the inverse of the price sensitivity of the market, (2) the average value added by all brands in the market, and (3) the value advantage (or disadvantage) of the brand. The weights depend primarily upon the number of competing brands, with price insensitivity having the strongest impact, followed by value advantage of the brand, and average value added by all brands. This approximation for optimal price is found to be robust under a wide range of conditions. Additionally, we demonstrate that using the approximation results in only marginal deviation of profits from the theoretical Nash optimal.     

间接和混合渠道下零售商引入自有品牌的影响分析

自有品牌的引入使渠道中制造商和零售商的市场绩效发生了较大变化。本文构建了两个制造商和一个零售商情境下零售商引入自有品牌的模型,分析了在间接渠道和混合渠道两种条件下引入自有品牌对零售商、制造商市场绩效的影响。通过分析表明,在间接渠道和混合渠道下,零售商引入自有品牌后,一方面会导致制造商品牌的批发价格、零售价格和制造商利润的降低,对制造商不利;另一方面能增加渠道总利润、零售商利润及零售商在整个渠道中的利润分成,零售商应该将自有品牌的战略定位为高质高价的品牌,而非低价低质的原始状态型品牌,而制造商单纯靠改变销售渠道模式的策略并不足以对抗自有品牌的冲击。     

家庭经济因素和人口特征如何影响不同收入等级城镇居民消费

本文使用2007年中国家庭收入项目(CHIPS)调查数据,在考虑家庭消费影响因素的线性和非线性效应的前提下,采用可加的半参数分位数回归模型,在不同消费水平上研究家庭经济因素和家庭人口特征是否影响消费.结果表明,消费水平较高的家庭的收入弹性也比较大.户主受教育年限和家庭子女数量对消费具有正向影响,尤其对低收入家庭影响显著.家庭消费在地区上存在明显差异,地区效应对中部低消费家庭的影响尤其显著.户主年龄、户主健康状况和家庭规模也是家庭消费的显著影响因素.     

一种电子商务环境下耐用品的动态定价模型与策略

针对电子商务环境下消费者对价格歧视的抗拒问题,以及耐用品生命周期长、产品需求依赖于时间、价格等特点,提出了一种动态定价模型与策略。该模型通过构造转移概率矩阵,推导出在线消费者浏览到耐用品的不同价格状态下的概率,接着根据消费者多阶段效用函数分析消费者的购买决策行为,进而给出零售商利润达到最大化时的最优定价策略集合。为了验证模型与策略的有效性,通过数值模拟实验,分析模型主要参数变化对最优定价策略的影响。研究发现当效用折扣因子越高,零售商应该降低促销频率和高价格并且提高低价格,从而诱导高端消费者在高价格购买产品。折扣效用因子大小还决定了网上零售商是否要隐藏自己的促销概率。     

Spatial mapping of price competition using logit-type market share models and store-level scanner-data

This paper proposes a methodology to obtain reliable spatial maps of price competition using store-level scanner data. Specifically, a procedure to obtain a symmetric matrix of similarities between brands considering their substitutability depending on price variations is proposed. The matrix is derived from a market response model where price cross-effects are split into two components. The first component accounts for the fact that price variation in one brand can have different effects to price variation in other brands (ie j → j′≠j′ → j). The second component accounts for the fact that the price of each brand can have different effects across competing brands (ie j → j′≠j → j^″). The matrix is obtained by imposing symmetry on this second component of price cross-effects. The parameterization of this symmetric matrix of similarities as the distances between the spatial representations of brands allows us to obtain the positioning maps. The proposed approach is illustrated through an empirical application.     

Collaborative pricing and replenishment policy for hi-tech industry

Owing to rapid technological innovation and severe competition, the upstream component price and the downstream product cost of hi-tech industries like computers and communication consumer's products usually decline significantly with time. From a practical viewpoint, there is a need to develop a collaborative pricing and replenishing model with finite horizon when the vendor's purchase cost and the end-consumer's market price are reduced simultaneously. To entice collaboration, the vendor may offer some price discount to the buyer using a negotiation factor to balance the net profit for each player. A numerical example and sensitivity analysis are carried out to illustrate the model. Our results indicate that higher decline-rate in the vendor's purchase cost leads to a smaller vendor lot size, and the higher decline-rate in the market price leads to a larger buyer lot size. The percentage increase in the net profit is approximately 6.57% when cost/price reduction is considered. Therefore, it is significant to consider the effect of the cost/price reduction, especially in hi-tech industries.     

Household consumption, investment and life insurance

This paper develops a continuous-time Markov model for utility optimization of households. The household optimizes expected future utility from consumption by controlling consumption, investments and purchase of life insurance for each person in the household. The optimal controls are investigated in the special case of a two-person household, and we present graphics illustrating how differences between the two persons affect the controls.     

Supply chains in the presence of store brands

Increased competition from store brands is forcing manufacturers to re-evaluate their strategies in regard to pricing and contracting with trade intermediaries. We analyze a supply chain in which a retailer accepts (with the appropriate contractual agreements) a national brand for resale and then determines whether to introduce a store brand, how to price the store brand, and what quantities of the product(s) to order. We show that when the national brand’s cost per unit quality (CPUQ) is larger than the store brand’s CPUQ, then the retailer seeks to introduce the store brand (SB) and the national brand (NB) manufacturer/supplier is unable to deter him from doing so. We find that the efficiency loss in the decentralized supply chain becomes smaller when a store brand is introduced. Recognizing the inadequacy of standard contracts in coordinating this supply chain, we propose a simple minimum order quantity contract that can coordinate this supply chain.     

National and store brand advertising strategies

As the propensity of premium store brands (SBs) increases, retailers must consider different ways to drive sales besides promotional strategies. With this in mind, we consider a national brand (NB) and a (premium) SB co-existing in a market. Each brand has to decide the amount to invest in advertising its product and the prices to charge its customers, which can be determined separately or in unison. When either advertising expenditures or pricing decisions are set, each brand must keep in mind that the advertising efforts and revenue may spillover between the two brands, customers who intend to purchase the NB may end up purchasing the SB and vice versa. We derive an analytical model of the situations described and characterize equilibrium advertising decisions. We find that the characteristics of a premium SB may depend on which marketing/promoting instrument (advertising or pricing) is the primary method for driving demand; and in some situations an NB may be better off to not advertise at all and instead let the premium SB carry out all of the advertising.