Unit 4 DB AMDM

profileSteelTiger4
LiveChat4studentnotes.ppt
Home>Business & Finance homework help>Marketing homework help>Unit 4 DB AMDM

APPLIED MANAGERIAL DECISION-MAKING
MGMT 600

AGENDA

  • Qualitative vs. Quantitative
  • Types of Variables

Independent and Dependent

  • Regression Analysis

Why

How

  • Samples of Regression Testing
  • Week 4 Assignments

RESEARCH REVIEW

  • Starts with the question
  • Establish a Hypothesis
  • Decide what variable you would like to test
  • Determine the type of data that you want to gather

*

TYPES OF DATA

  • Qualitative research gathers information that is not in _________ form.

For example, diary accounts, open-ended questionnaires, unstructured interviews and unstructured observations. 

Qualitative data is typically ___________ data and as such is harder to analyze than quantitative data.

*

TYPES OF DATA

  • Quantitative data is data expressing a certain quantity, amount or range. Usually, there are ____________ units associated with the data, e.g. _____ and _____, in the case of the height of a person.

It makes sense to set _________limits to such data, and it is also meaningful to apply arithmetic operations to the data.

For example, measurements, weights, values, or any gathered information in numerical form.

*

USE OF DATA

  • Qualitative and Quantitative data is used to …

Evaluate the relationship between a dependent variable and at least one independent variable.

A dependent variable is the variable being tested and measured _____________________.

An __________________ is the variable that is changed or controlled in a scientific experiment to test the effects on ___________________

*

TYPES OF DATA

QUALITATIVE

Is it too hot or old?

Is your tea too sweet?

Is an item too large or small?

Did you improve?

SDisagree Disagree Neutral Agree SAgree

QUANTITATIVE

Degrees

Amount of sugar or sweetener

Actual size

Measured change

Actual Value

REGRESSION CORRELATIONS

*

  • ________ or _________ relationships
  • Correlation does not imply ____________
  • Regression is denoted by an “r”
  • Slope of the line (negative or positive) is determined by calculating coefficient and t-stat (demo coming)

REGRESSION CORRELATIONS

*

Regression analysis

This technique is used

Regression analysis is

REGRESSION MODELING

*

Linear Regression -

Logistic Regression -

Polynomial Regression –

Stepwise Regression –

Ridge Regression –

Lasso Regression –

ElasticNet Regression -

REGRESSION MODELING

*

MULTIVARIATE

  • Multivariate statistics

In order to be meaningful, market survey questions must often be complex and multilayered and answers can vary depending on a number of conditions and factors.

Business people use multivariate techniques to analyze consumer responses by grouping the different consumer responses into meaningful categories or clusters of information in order to determine the relevant relationships between the responses and the product or service being studied.

*

MULTIVARIATE

  • Factors Influencing the selection of a Data Analysis Strategy

Known characteristics of the data: For example, scale of the measurement (ordinal, interval or ratio). Certain techniques are only appropriate for certain types of data.

*

CHOICE OF METHODOLOGY

  • Classification of Statistical Techniques

Statistical technique can be univariate or multivariate

_________ are appropriate for analyzing data when there is a single measurement for each element in the sample

Or if there are several measurements of each variable, each variable is analyzed in isolation.

Example: Chi-square

*

CHOICE OF METHODOLOGY

  • Key Difference

Multivariate statistical techniques can be classified as dependent or independent techniques.

__________ techniques can be useful when one or more variables can be identified as dependent and the rest as independent

Example: Conjoint Analysis, Multiple Regression, Discriminate Analysis

___________ techniques, variables are not classified as dependent or independent, rather the whole set of interdependent relationships is examined.

Example: Factor Analysis, Cluster Analysis, Multidimensional Scaling

*

UNIVARIATE: EXAMPLE

  • CONJOINT ANALYSIS

Conjoint analysis looks for a preference for a combination of product characteristics in order to predict individual preferences. It is often referred to as “______-________.” For instance, if a consumer states that her priorities when shopping for shoes is comfort, style, price and availability, the marketer might conclude that comfort and style are the most important. But the other factors of price and availability might rank higher if considered together, so that price and availability together might be more important than style.

*

UNIVARIATE: EXAMPLE

  • MULTIPLE REGRESSION

Multiple regression analysis is the most commonly used multivariate technique. It is used as a ________ tool to analyze the relationship between a single __________ response, such as whether the consumer prefers a certain product such as a brand of ice cream, and two or more independent responses, such as the person’s age and income.

*

UNIVARIATE: EXAMPLE

  • Discriminate Analysis

Discriminant analysis is used to create _____________ of people based on a variety of characteristics, such as whether the person is an “outdoors” person or do-it-yourselfer, and then determine which classification the person being surveyed fits into based on how his responses match up.

*

MULTIVARIATE: EXAMPLE

  • Factor Analysis

Purpose of factor analysis is to describe the covariance _________ among many variables in terms of underlying but unobservable, random qualities called factors.

If each variable could be grouped by their ____________(i.e. shopping patterns or recreational activities, or food preferences) then it is conceivable that each group represents a single _________.

*

MULTIVARIATE: EXAMPLE

  • Cluster Analysis

Cluster analysis also reduces a large __________ into a small number of mutually exclusive groups based on similar characteristics. Examples would be “baby boomers,” “geeks” or “yuppies.”

Cluster analysis is an exploratory analysis that tries to identify structures within the data.  More specifically, it tries to identify ____________ groups of cases if the grouping is not previously known.  Because it is exploratory, it does not make any distinction between ___________ and independent variables. 

*

MULTIVARIATE: EXAMPLE

  • Multidimensional Scaling

Can be considered an ________ to factor analysis

Multidimensional scaling ____ consumer responses on a graph according to similarity so that different responses can be analyzed based on their _________ to other responses.

Goal of the analysis is to detect meaningful underlying dimensions that allow the researcher to explain __________ similarities or dissimilarities (distances) between the investigated objects

*