Saturday, 15 September 2012

15 sept team b


Blog

Factor analysis is commonly used in data reduction, scale development, psychometric quality of a measure and the assessment of the dimensionality of a set of variables. Although it can be used for a variety of purposes it is used to determine a small number of factors based on a particular number of inter-related quantitative variables. Unlike directly measured variables, some variables are not a single measurable entity. They are constructs that are derived from the measurement of other, directly observable variables. Factor analysis measures not directly observable constructs by measuring several of its underlying dimensions. The identification of such underlying factors simplifies the understanding and description of complex constructs.
From this we can infer that factor analysis reduces the multitude of observable factors into a smaller set of factors and then these factors can be used to explain complex phenomenon using tools like permap etc.From the use of the various tools later we can conclude or observe relationships between the various factors which in turn are relationships between the directly observable factors and then it can be turned to strategy.
Then we proceed to discriminant analysis. It is a feature of SPSS which allows us to build predictive models for grouping .It is composed of a discriminant function based on linear combination of predictor variables. The purpose of this is to group variables and to discard variables which are little related to the group distinctions. We would like to know the relationship between a group of independent variables and one category variable. Using this relationship we can predict a classification based on the independent variables or we can also evaluate how well the independent variables separate the categories in the classification.
 It is similar to regression analysis. A discriminant score can be calculated from the weighted combination of the independent variables. We then use a maximum likelihood technique to assign a case to a group from a specified cut off score.  We used this method to predict the defaulters of a bank loan and tested the theory with the SPSS software. By using discriminant analysis we could develop a model to find the number of people who would be defaulters and decide whether to provide the loan to them or not.


By Rohith Emmanuel
Group B

No comments:

Post a Comment