Cluster analysis versus Q-type factor analysis as a disaggregation method in hybrid conjoint modeling: An empirical investigation

The author compares the predictive performance of cluster analysis and Q-type factor analysis as disaggregation methods in hybrid conjoint modeling. The results indicate that cluster- and Q-type-factor-based models perform about equally well in predictive validity terms. The paper, however, suggests the need for more empirical research.     

An empirical test of optimal respondent weighting in conjoint analysis

This article empirically tests a proposal (due to Hagerty) to use Q-type factor analysis to maximize predictive accuracy in conjoint analysis. Data sets from three different studies are used to compare the accuracy of predictions from optimal weighting with those from individual conjoint parameter estimation. The results do not support the contention that optimal weighting significantly improves cross-validity, as compared to individual conjoint prediction. He has been honored for his research by the American Marketing Association, the American Statistical Association, the American Psychological Association, and the Market Research Society (England). He has authored or coauthored several books, including the widely used textResearch for Marketing Decisions, now in its fifth edition. He is also a prolific contributor to marketing and business journals. He is the author or coauthor of many articles in statistical methodology and the interface between statistical methodology and optimization theory. His current research interests include theoretical and empirical analyses of the bootstrap resampling technique and application of statistical methods and operations research to problems in marketing research. She received her Ph.D. in marketing from Drexel University in 1990. Her research interests focus on consumer preference models including conjoint analysis and new-product development. Her work has been published in theJournal of Marketing Research, Journal of Advertising Research, andJournal of the Market Research Society.     

A test of the convergent validity of self-explicated and decompositional conjoint measurement

An experiment was conducted to assess the convergent validity of a recently proposed conjunctive-compensatory self-explicated approach to conjoint measurement. A comparison of the self-explicated approach with the full profile conjoint method demonstrated that the self-explicated approach (1) was inconsistent in the identification of acceptable and unacceptable stimuli, and (2) didnot exhibit a high degree of convergent validity with respect to partworth and importance estimates when compared with the full profile approach. These findings identify potential problems with the conjunctive-compensatory self-explicated approach to conjoint measurement.     

Predictive performance of self-explicated,traditional conjoint,and hybrid conjoint models under alternative data collection modes

Although data collection in conjoint analysis has traditionally involved the use of in-person interviews, recent years have seen a trend toward the use of alternative data collection modes, including mail questionnaires and telephone interviews. Since these alternative modes differ in environment, the author examines the predictive performance of conjoint models under three data collection modes, i.e., in-person interviews, mail questionnaire, and telephone interviews. The results indicate the conjoint models examined to be comparable in predictive, performance across the three data collection modes.     

Collecting conjoint data through telephone interviews

Data collection for conjoint analysis is relatively new. The current study examines the reliability and convergent validity for selected user segments vs. the mail questionnaire alternative. The results attest the telephone collection procedure is both reliable and valid relative to collecting data with mail questionnaires.     

Preference mapping of conjoint-based profiles: An INDSCAL approach

This article describes an approach in which conjoint methodology and multidimensional scaling can be fruitfully applied, in tandem, to the measurement and representation of preference data for factorially designed profiles. The approach is described and applied to an illustrative data set. We conclude the article with a brief discussion of possible extensions of the approach and additional research needs. AT&T Bell Laboratories     

Examining the effects of regression procedures on the temporal stability of parameter estimates in marketing models

This article investigates the temporal stability of parameter estimates by comparing the results obtained from ordinary least squares, ridge, and latent root regression techniques in two models plagued with ill-conditioned data. Ridge regression is found to provide improved individual coefficient stability through time and also to have slightly greater predictive accuracy beyond the original estimation period in both models. Temporal stability of marketing model parameters is important if the model is to prove useful to marketing decision makers in making inferences about the marginal impacts of individual predictor variables.     

某变电站主控楼、配电室结构检测与鉴定

文章介绍了某大型变电站的主控楼和配电室的结构可靠性检测鉴定。对地基基础采用了经纬仪进行倾斜测量,结果显示该建筑物主体结构目前无明显倾斜和不均匀沉降,表明主体结构的基础现状已基本稳定。对于上部承重结构,框架柱、梁的混凝土强度,采用回弹法进行了抽查检测,检测结果显示主控楼、配电室结构构件现状的混凝土强度等级可以满足设计要求。实践证明该方法可为此类建筑物的结构鉴定提供一定的技术参考。     

Measuring core inflation in Italy comparing aggregate vs. disaggregate price data

This paper focuses on the core inflation measurement in Italy using univariate (national-level inflation) vs. multivariate (city-level inflation) models during the period 1970–2006. We derive algebraic expressions that allow comparison between the reduced form parameters of univariate and multivariate local level models in the context of contemporaneous and temporal aggregation. We illustrate the relevance of these theoretical results for the empirical analysis of time series. Using Italian data, we find that multivariate and univariate models extract similar core inflation measures when analyzing the moderate-low inflation period. In contrast, the two competing models yield different trends when modeling the Great Inflation period.     

旅游景区形象对顾客忠诚的影响研究

采用文献回顾和定性研究相结合的方法,提出了旅游景区形象4维度测量模型。基于此,建立了旅游景区形象各维度对顾客忠诚的作用机制模型．数据研究表明：（1）笔者开发的旅游景区形象量表具有较好的信度水平、整体拟合优度、收敛效度、判别效度,它包括风景与卫生、景区价格、景区服务、基础设施等4个结构维度;（2）旅游景区形象各维度对顾客忠诚存在不同的作用机理与影响效应．     

Choice rules and sensitivity analysis in conjoint simulators

Despite the widespread use of choice simulators in commercial conjoint applications, relatively little has been written about the applicability of various types of buyer choice rules and sensitivity analyses. This paper first discusses issues related to the selection of different buyer choice rules. We then propose six types of sensitivity analyses that can be implemented in buyer choice simulators, given the usual input data of respondents’ part worths, status quo product utilities, and background data. Each sensitivity analysis is illustrated in the context of a common business application. The paper concludes with a brief discussion of possible extensions of sensitivity analysis and areas for further research.     

The dual role of price: decomposing consumers’ reactions to price

Price plays two distinct roles in consumers’ evaluations of product alternatives: as a measure of sacrifice and as an informational cue. This article merges two streams of empirical research into the effects of price on consumers’ product evaluations by combining stated preferences, obtained from conjoint measurement, with data on self-reported measures in the form of beliefs or attitudes. It thus offers new, substantive insights into the dual role of price. Specifically, it differentiates between the informational and sacrifice effects of price using a choice-based conjoint approach and differentiates further among different subcomponents of these two main effects by combining choice-based measures with self-reported measures that pertain to potential sources of the dual role of price (price response drivers) and underlying consumer characteristics. Thus, this article presents a general procedure to quantify the impact of the dual role of price on choice shares for product alternatives within a market simulation. This procedure enables managers to simulate the choice share effects of changes in price response drivers, as well as modifications in segmentation and targeting strategies that involve changes in the levels of the price response drivers and thus the levels of the informational and sacrifice components of the price response of demand.     

Mismeasuring long-run growth: the bias from splicing national accounts—the case of Spain

Comparisons of economic performance over space and time largely depend on how statistical evidence from national accounts and historical estimates are spliced. To allow for changes in relative prices, GDP benchmark years in national accounts are periodically replaced with new and more recent ones. Thus, a homogeneous long-run GDP series requires linking different temporal segments of national accounts. The choice of the splicing procedure may result in substantial differences in GDP levels and growth, particularly as an economy undergoes deep structural transformation. An inadequate splicing may seriously bias the measurement of GDP levels and growth rates. Alternative splicing solutions are discussed in this paper for the particular case of Spain, a fast-growing country in the second half of the twentieth century. It is concluded that the usual linking procedure, retropolation, is seriously flawed as it tends to bias GDP levels upwards and, consequently, to underestimate growth rates, especially for developing countries experiencing structural change. An alternative interpolation procedure is proposed.     

基于SPIDER法测量飞秒脉冲的数值分析与模拟

阐述了目前国际上公认的用于飞秒脉冲测量的光谱相位相干电场重构法(SPIDER)的光谱图数值分析过程，并对高斯强度分布的变换极限脉冲、线性啁啾脉冲进行数值计算及模拟，都获得了准确还原的飞秒脉冲在频域和时域的强度包络和相位．     

The effect on sales of changes in a “push” marketing strategy in a marketing channel context

Marketing managers must determine what level of salient marketing mix variables should be provided within a marketing channel. This paper describes a number of previously used methods and presents a different approach which determines a customer's sales response to different levels of promotion and distribution activities using a variant conjoint analysis approach. The results of this sales dollar estimation procedure are compared with those derived from conventional conjoint analysis using rank order preference data. Differences in the two analyses are examined which could lead to different strategic decisions.     

Clustering consistency analysis

Cluster analysis is a frequently used technique in marketing as a method to develop partitions or classifications for market segmentation, product positioning, test market selection, etc. Because of the vast diversity in the assortment of clustering algorithms available, it is often times not obvious which algorithm or technique should be employed. It is often recommended that the marketer perform more than one cluster analysis on the same data set and compare representations as a reliability check. A methodology for evaluating the consistency of different clusterings is introduced via contingency table analysis by log-linear models. In addition, insight is provided as to selecting a “best” representative clustering by examining Stewart and Love's (1968) redundancy measures. Bell Laboratories     

Magnitude scaling of the dependent variable in decompositional multiattribute preference models

Conclusion The results of this study produced strong support for the validity magnitude scaling procedures when aggregate data are analyzed. For example, the cross modality matching procedures indicated the geometric means of the preference responses to the CMA stimuli were characterized by power functions with estimated exponents similar to the theoretically derived exponents. When the data were analyzed at the individual level, however, considerable heterogeneity across individuals was found--the power function exponents vary considerably from individual to individual. This heterogeneity across subjects is similar to the findings of previous research on magnitude scaling (Dawson 1982). This may, therefore be a major limitation of this scaling procedure in conjoint analysis applications because of the widespread use of conjoint analysis for the examination of individual respondent utility functions.     

An exploratory study of the generalizability of selected marketing measures

Generalizability theory, a broad-based psychometric model for assessing the dependability (reliability) of measures, is used to assess several commonly used marketing measures. Classical reliability coefficients are reported for each measure and compared to generalizability coefficients. Classical theory is shown to be inappropriate in some marketing contexts and the marketing measures are found to be undependenable in some measurement situations. The comprehensiveness and flexibility of generalizability theory is demonstrated.     

关于教育测量的要素分析

教育测量是指对教育现象的测量。教育测量的基本手段是存在失真范围的数量化分析。其数量化的语言是效度、信度等。效度分为效标效度与内容效度等,信度是表明测量可靠性的数量指标。