• Embed Doc
  • Copy Link
  • Readcast
  • Collections
  • CommentGo Back
Download
 
MARKETRESEARCH
SubmittedBy:Himanshu Arora 06Hemali Chawla 18Nitin Chawla 19
 
Multivariate AnalysisTechniques:
Situation 1:
A harried executive walks into your office with a stack of printouts. She says,“You’re the marketing research whiz—tell me how many of this new redwidget we are going to sell next year. Oh, yeah, we don’t know what pricewe can get for it either.”
Situation 2:
Another harried executive (they all seem to be that way)calls you into his office and shows you three proposed advertisingcampaigns for next year. He asks, “Which one should I use? They all lookpretty good to me.”
Situation 3:
During the annual budget meeting, the sales manager wantsto know why two of his main competitors are gaining share. Do they havebetter widgets? Do their products appeal to different types of customers?What is going on in the market?All of these situations are real, and they happen every day acrosscorporate. Fortunately, all of these questions are ones to which solid,quantifiable answers can be provided. An astute marketing researcherquickly develops a plan of action to address the situation. The researcherrealizes that each question requires a specific type of analysis,Over the past 20 years, the dramatic increase in desktop computingpower has resulted in a corresponding increase in the availability of computation intensive statistical software. Programs like SAS and SPSS,once restricted to mainframe utilization, are now readily available inWindowsbased, menu-driven packages. The marketing research analystnow has access to a much broader array of sophisticated techniques withwhich to explore the data. The challenge becomes knowing whichtechnique to select, and clearly understanding their strengths andweaknesses.
WHAT IS MULTIVARIATE ANALYSIS?
Multivariate analysis is the analysis of the simultaneous relationshipsamong three or more phenomena. While in a univariate analysis the focusis on the level (average) and distribution (variance) of the phenomenon,while in a bivariate analysis the focus shifts to the degree of relationships(correlations or covarainces) between the phenomenon. In a multivariateanalysis, the focus shifts from paired relationships to the more complexsimultaneous relationships among phenomenon.Multivariate analysis (MVA) is based on the statistical principle of multivariate statistics, which involves observation and analysis of more
 
than one statistical variable at a time. In design and analysis, thetechnique is used to perform trade studies across multiple dimensionswhile taking into account the effects of all variables on the responses of interest.Uses for multivariate analysis include:* Design for capability (also known as capability-based design)* Inverse design, where any variable can be treated as an independentvariable* Analysis of alternatives, the selection of concepts to fulfill a customerneed* Analysis of concepts with respect to changing scenarios* Identification of critical design drivers and correlations acrosshierarchical levelsMultivariate analysis can be complicated by the desire to include physics-based analysis to calculate the effects of variables for a hierarchical"system-of-systems." Often, studies that wish to use multivariate analysisare stalled by the dimensionality of the problem. These concerns are ofteneased through the use of surrogate models, highly accurateapproximations of the physics-based code. Since surrogate models takethe form of an equation, they can be evaluated very quickly. Thisbecomes an enabler for large-scale MVA studies: while a Monte Carlosimulation across the design space is difficult with physics-based codes, itbecomes trivial when evaluating surrogate models, which often take theform of response surface equations.
Decision Analyst 
In order to understand multivariate analysis, it is important to understandsome of the terminology. A variate is a weighted combination of variables. The purpose of the analysis is to find the best combination of weights.Nonmetric data refers to data that are either qualitative or categorical innature. Metric data refers to data that are quantitative, and interval orratio in nature.
Initial Step—Data Quality
Before launching into an analysis technique, it is important to have a clearunderstanding of the form and quality of the data. The form of the datarefers to whether the data are nonmetric or metric. The quality of the datarefers to how normally distributed the data are. The first few techniquesdiscussed are sensitive to the linearity, normality, and equal varianceassumptions of the data. Examinations of distribution, skewness, andkurtosis are helpful in examining distribution. Also, it is important tounderstand the magnitude of missing values in observations and todetermine whether to ignore them or impute values to the missingobservations. Another data quality measure is outliers, and it is important
of 00

Leave a Comment

You must be to leave a comment.
Submit
Characters: ...
You must be to leave a comment.
Submit
Characters: ...