The author presents the general EM algorithm for analyzing incomplete data. As a specific application of the EM algorithm, a model is proposed for analyzing incomplete data in an important class of problems in marketing research. A simple estimation procedure also is developed. The model is investigated through Monté Carlo studies as well as empirically with encouraging results. Some of the advantages and limitations of the approach are discussed.
Get full access to this article
View all access options for this article.
References
1.
AkaahIshmael, and KorgaonkarPradeep K. (1983), “An Empirical Comparison of the Predictive Validity of Self-Experienced, Huber-Hybrid, Traditional Conjoint, and Hybrid Conjoint Models,”Journal of Marketing Research, 20 (May), 187–97.
2.
AmemiyaTakeshi (1973), “Regression Analysis When the Dependent Variable is Truncated Normal,”Econometrica, 41 (November), 997–1016.
3.
AmemiyaTakeshi (1975), “Qualitative Response Models,”Annals of Economic and Social Measurement, 4 (Summer), 363–72.
4.
AmemiyaTakeshi (1979), “The Estimation of a Simultaneous Equation Tobit Model,”International Economic Review, 20 (February), 169–81.
5.
AndersonAndy B., BasilevskyAlexander, and HumDerek P. J. (1983), “Missing Data,” in Handbook of Survey Research, RossiPeter H., WrightJames D., and AndersonAndy B., eds. Orlando: Academic Press, Inc.,415–94.
6.
BoylesRussell A. (1983), “On the Convergence of the EM Algorithm,”Journal of the Royal Statistical Society, 45B, 47–50.
7.
CarmoneFrank J., and GreenPaul E. (1981), “Model Misspecifications in Multiattribute Parameter Estimation,”Journal of Marketing Research, 17 (February), 87–93.
8.
CarpenterGregory S., and LehmannDonald R. (1985), “Model of Marketing Mix, Brand Switching, and Competition,”Journal of Marketing Research, 22 (August), 318–29.
9.
CarterW. M., and MyersR. H. (1973), “Maximum Likelihood Estimation from Linear Combinations of Discrete Probability Functions,”Journal of the American Statistical Association, 68 (March), 203–6.
10.
ChowWinston K. (1979), “A Look at Various Estimators in Logistic Models in the Presence of Missing Values,”American Statistical Association Proceedings, Business Economics Section, 417–20.
11.
CohenJacob, and CohenPatricia (1975), Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences.New York: John Wiley & Sons, Inc.
12.
CronWilliam L., and SlocumJohn W.Jr. (1986), “The Influence of Career Stages on Salespeople's Job Atittudes, Work Perceptions, and Performance,”Journal of Marketing Research, 23 (May), 119–29.
13.
DagenaisM. G. (1976), “Incomplete Observations and Simultaneous Equations Models,”Journal of Econometrics, 4 (August), 231–42.
14.
DahiyaR. C., and KorwarR. M. (1980), “Maximum Likelihood Estimates for Bivariate Normal Distribution with Missing Data,”Annals of Statistics, 8 (May), 687–92.
15.
DempsterA. P., LairdN. M., and RubinD. B. (1977), “Maximum Likelihood from Incomplete Data via the EM Algorithm,”Journal of the Royal Statistical Society, 39B, 1–22.
16.
DeSarboWayne S. (1978), “MEMD: An APL Program for the Multivariate Estimation of Missing Data,”Journal of Marketing Research, 15 (February), 117–8.
17.
DeSarboWayne S., GreenPaul E., and Douglas CarrollJ. (1986), “An Alternating Least-Squares Procedure for Estimating Missing Preference Data in Product Concept Testing,”Decision Sciences, 17 (Spring), 163–85.
18.
DunsmuirW., and RobinsonP. M. (1981), “Parametric Estimators for Stationary Time Series With Missing Observations,”Advanced Applied Problems, 13, 129–46.
19.
FinkbeinerCarl (1979), “Estimation for the Multiple Factor Model When Data are Missing,”Psychometrika, 44 (December), 409–20.
20.
FraneJames W. (1976), “Some Simple Procedures for Handling Missing Data in Multivariate Analysis,”Psychometrika, 41 (September), 409–15.
21.
GleasonTerry C., and StaelinRichard (1975), “A Proposal for Handling Missing Data,”Psychometrika, 40 (June), 229–52.
22.
GreenPaul E. (1974), “On the Design of Choice Experiments Involving Multifactor Alternatives,”Journal of Consumer Research, 1 (September), 61–8.
23.
GreenPaul E., GoldbergStephen M., and MontemayorMila (1981), “A Hybrid Utility Estimation Model for Conjoint Analysis,”Journal of Marketing, 45 (Winter), 33–41.
24.
GreenPaul E., and SrinivasanV. (1978), “Conjoint Analysis in Consumer Research: Issues and Outlook,”Journal of Consumer Research, 5 (June), 103–23.
25.
GreeneWilliam H. (1981), “On the Asymptotic Bias of the OLS Estimator of the TOBIT Model,”Econometrica, 49 (March), 505–13.
26.
HabermanS. J. (1977), “Discussion on the Paper by Professor Dempster, Professor Laird and Dr. Rubin,”Journal of the Royal Statistical Society, 39B, 31–2.
27.
HartleyH. O. (1958), “Maximum Likelihood Estimation from Incomplete Data,”Biometrics, 14 (June), 174–94.
28.
HartleyH. O., and HockingR. R. (1971), “The Analysis of Incomplete Data,”Biometrics, 27 (December), 783–808.
29.
HartleyMichael J. (1976), “The Tobit and Probit Models: Maximum Likelihood Estimation by Ordinary Least Squares,” working paper, State University of New York at Buffalo.
30.
HeckmanJames J. (1976), “The Common Structure of Statistical Models of Truncation, Sample Selection and Limited Dependent Variables and a Simple Estimator for Such Models,”Annals of Economic and Social Measurement, 5 (Fall), 475–92.
31.
HeckmanJames J. (1979), “Sample Bias as a Specification Error,”Econometrica, 47 (January), 153–62.
32.
HockingR. R., and MarxD. L. (1979), “Estimation with Incomplete Data: An Improved Computational Method and the Analysis of Nested Data,”Communications in Statistics: Theory and Methods, 8A, 1155–82.
33.
HockingR. R., and OxspringH. H. (1971), “Maximum Likelihood Estimation with Incomplete Multinomial Data,”Journal of the Royal Statistical Society, 66B, 65–70.
34.
KimJae-On, and CurryJames (1977), “The Treatment of Missing Data in Multivariate Analysis,”Sociological Methods and Research, 6, 215–60.
35.
LairdNan (1978), “Nonparametric Maximum Likelihood Estimation of a Mixing Distribution,”Journal of the American Statistical Association, 73 (December), 805–11.
36.
de LignyC. L., NieuwdorpG. H. E., BrederodeW. K., HammersW. E., and Van HouwelignenJ. C. (1981), “An Application of Factor Analysis with Missing Data,”Technometrics, 23 (February), 91–5.
37.
LittleR. J. A. (1977), “Discussion on the Paper by Professor Dempster, Professor Laird and Dr. Rubin,”Journal of the Royal Statistical Society, 39B, 25.
38.
LittleR. J. A. (1978), “Consistent Regression Methods for Discriminant Analysis with Incomplete Data,”Journal of the American Statistical Association, 73 (June), 319–22.
39.
LittleR. J. A. (1979), “Maximum Likelihood Influence for Multiple Regression with Missing Values: A Simulation Study,”Journal of the Royal Statistical Society, 41B, 76–87.
40.
MadowWilliam G., NisselsonHarold, and OlkinIngram (1983), Incomplete Data in Sample Surveys, Vol. 1. New York: Academic Press, Inc.
41.
MadowWilliam G., OlkinIngram, and RubinDonald B. (1983), Incomplete Data in Sample Surveys, Vol. 2. New York: Academic Press, Inc.
42.
MalhotraNaresh K. (1982a), “Information Load and Consumer Decision Making,”Journal of Consumer Research, 8 (March), 419–30.
43.
MalhotraNaresh K. (1982b), “Structural Reliability and Stability of Nonmetric Conjoint Analysis,”Journal of Marketing Research, 19 (May), 199–207.
44.
MalhotraNaresh K. (1983), “A Comparison of the Predictive Validity of Procedures for Analyzing Binary Data,”Journal of Business and Economic Statistics, 1 (October), 326–36.
45.
MalhotraNaresh K. (1984), “The Use of Linear Logit Models in Marketing Research,”Journal of Marketing Research, 21 (February), 20–31.
46.
MalhotraNaresh K. (1986a), “Modeling Store Choice Based on Censored Preference Data,”Journal of Retailing, 62 (Summer), 128–44.
47.
MalhotraNaresh K. (1986b), “An Approach to the Measurement of Consumer Preferences Using Limited Information,”Journal of Marketing Research, 23 (February), 33–40.
48.
OlsenRandall J. (1980), “Approximating a Truncated Normal Regression with the Method of Moments,”Econometrica, 48 (July), 1099–1105.
49.
PatelH. I., and KhatriC. G. (1981), “Analysis of Incomplete Data in Experiments with Repeated Measurements in Clinical Trials Using a Stochastic Model,”Communications in Statistics: Theory and Methods, 10, 2259–78.
50.
PunjGirish N., and StaelinRichard (1978), “The Choice Process for Graduate Business Schools,”Journal of Marketing Research, 15 (November), 588–98.
51.
RaoRam C., and BassFrank M. (1985), “Competition, Strategy, and Price Dynamics: A Theoretical and Empirical Investigation,”Journal of Marketing Research, 22 (August), 283–96.
52.
RobinsonWilliam T., and FornellClaes (1985), “Sources of Market Pioneer Advantages in Consumer Goods Industries,”Journal of Marketing Research, 22 (August), 305–17.
53.
SchmeeJoseph, and HahnGerald J. (1979), “A Simple Method for Regression Analysis with Censored Data,”Technometrics, 21 (November), 417–32.
54.
SchwertmanN. C., FridshalD., and MagreyJ. M. (1981), “On the Analysis of Incomplete Growth Curve Data: A Monte Carlo Study of Two Nonparametric Procedures,”Communications in Statistics: Simulation and Computation, 10B, 1, 51–66.
55.
SmithPatricia L. (1981), “The Use of Analysis of Covariance to Analyze Data from Designed Experiments with Missing or Mixed-up Values,”Applied Statistics, 30, 1–8.
56.
SrinivasanV., JainArun K., and MalhotraNaresh K. (1983), “Improving Predictive Power of Conjoint Analysis by Constrained Parameter Estimation,”Journal of Marketing Research, 20 (November), 433–8.
57.
SrinivasanV., and ShockerAllan D. (1981), “LINMAP Version IV-User's Manual,”Owen Graduate School, Vanderbilt University.
58.
SundbergR. (1974), “Maximum Likelihood Theory for Incomplete Data from an Exponential Family,”Scandinavian Journal of Statistics, 1, 49–58.
59.
SundbergR. (1976), “An Iterative Method for Solution of the Likelihood Equations for Incomplete Data from Exponential Families,”Communications in Statistics: Simulation and Computations, 5, 55–64.
60.
TobinJames (1958), “Estimation of Relationships for Limited Dependent Variables,”Econometrica, 26 (January), 24–36.
61.
WilliamsE. R., RatcliffD., and SpeedT. P. (1981), “Estimating Missing Values in Multi-Stratum Experiments,”Applied Statistics, 30, 71–2.
62.
WinerRussell S. (1983), “Attrition Bias in Econometric Models Estimated with Panel Data,”Journal of Marketing Research, 20 (May), 177–86.
63.
WittinkDick R., and CattinPhilippe (1981), “Alternative Estimation Methods for Conjoint Analysis: A Monté Carlo Study,”Journal of Marketing Research, 17 (February), 101–6.
