Sas - Bi With Sas
This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA
Overview
Download & View Sas - Bi With Sas as PDF for free.
More details
- Words: 772
- Pages: 3
Business Intelligence with SAS Program
: PGPBA
Class of
: 2009
Semester
:
Sessions
: 33
Credit
: 3
Course Code : IT661
Course Objective •
To provide concepts & techniques of Data Mining Analysis Tools (DMAT) which are different from various statistical techniques.
•
To equip the students with skills to perform data analysis and conclusions independently with special focus on Data Mining (DM) applications with SAS Enterprise Miner. REFERENCE BOOKS
AUTHOR / PUBLICATION
Data Mining Techniques: For Marketing, Sales, and Customer Relationship Management
Berry and Linoff - Wiley Computer Publishing; 2 Ed, 2004.
Applied Multivariate Techniques
Sharma Subhash - John Wiley & Sons, 1996
Using Multivariate Statistics’
Tabachnick B.G. & Fidell L.S - Allyn & Bacon, 1996.
Multivariate Data
Hair J.F, Anderson R.E., Tatham R.L, Black W.C - Hair. Pearson Education, 2003.
Detailed Syllabus
examples of data screening – Outlier analysis Residual analysis – Generalized linear models – types of sum of squares type I II III IV – Concept of AIC & SBC - Data partition into Training validation and testing
Introduction to Data Mining Applications in Marketing and Customer Relationship Management: A Statistical perspective of Data Mining - Analytic Customer Relationship Management - Tasks performed with data mining - Virtuous Cycle of Data Mining Applications of data mining – Concept of Learning , Knowledge discovery, Analytical Intelligence, Enterprise Intelligence
Receiver operating characteristics (ROC) : Concept , Construction and inferences – Area under the curve (AUC) – Multiclass problems – Volume under the curve(VUC) Logistic regression: Concept of odds ratio
Examining data using SAS: Testing the assumptions of multivariate analysis- BLUE Assessing individual variables Vs variate Normality – Heteroscedasticity, autocorrelation & Multicollinearity – Identification and solutions -absence of correlated errors Important issues in data screening - Complete PG Program in Business Administration
- Wald’s confidence interval and construction – concept of Cordant discordant and tied pairs Basic concepts of logistic regression - Logistic regression with only one categorical variable Logistic regression and contingency table analysis - Logistic regression for combination of 1
Class of 2008
categorical and continuous independent variables – Stepwise backward and forward regression methods in multivariate logistic regression Comparison of logistic regression
Rules: Concept of support confidence lift and gain Defining Market Basket Analysis - Three Levels of Market Basket Data - Order Characteristic - Item Popularity - Tracking Marketing Inventories - Clustering Product Usage - Association Rules - Actionable Rules Trivial Rules - Inexplicable Rules - Building Association Rules - Choosing the Right Set Of Items - Product Hierarchies Help to Generalize Items - Virtual Items Go Beyond the Product Hierarchies - Data Quality Anonymous Versus Identified - Generating Rules From All the Data - Calculating Confidence - Calculating Lift The Negative Rule - The Problem of Big Data
and discriminant analysis Decision Trees: Introduction - Growing a decision tree - concept of logworth – algorithms chaid & cart importance of variable selection Test for choosing the best split – Pruning Extracting rules from trees - Alternate representations for decision trees - Decision trees in practice Artificial Neural Networks: History - Real Estate Appraisal – Concept of a link function Neural Networks for Directed data Mining Neural Net - The Unit of a Neural Network Feed-Forward Neural Networks - Back Propagation Heuristics for using Feed-Forward Back Propagation Networks - Choosing the Training Set - Coverage of Values for All Features - Number of Features - Size of Training Set - Number of Outputs - Preparing the Data - Features with Continuous Values Features with Ordered, Discrete (Integer) Values - Features with categorical values - Other Types of Features - Interpreting the results - Neural Networks for Time series - Example: Finding Clusters - Lessons Learned
Suggested Schedule of Sessions Topic
Time Series Forecasting: Stationarity non stationarity Exponential Smoothing - ARIMA Models - AR Process - Moving Average Process - ARMA Process - Box Jenkins Methodology – Time series construction of rare events SAS Programming: Basics of Programming Data input through programming - Data Steps and Proc steps Simple SAS Programming – Construction of charts and plots using SAS steps
No of Sessions
Introduction
1
Examing Data (Enterprise Guide)
3
Exercises on Examing Data
2
Logistic Regression (LR) (Enterprise Guide)
3
Exercises on LR
2
Decision Trees (DT) (Enterprise Miner)
3
Exercise on DT
2
Artificial Neural Networks (ANN) (Enterprise Miner)
4
Exercise on ANN
2
Time Series Forecasting (TSFS)
3
Exercise on TSFS
1
SAS Programming (BASE SAS)
3
Market basket Analysis (MBA)(Enterprise Guide)
3
Exercise on MBA
1
Total Sessions
33
Market Basket Analysis and Association PG Program in Business Administration
2
Class of 2008
PG Program in Business Administration
3
Class of 2008
: PGPBA
Class of
: 2009
Semester
:
Sessions
: 33
Credit
: 3
Course Code : IT661
Course Objective •
To provide concepts & techniques of Data Mining Analysis Tools (DMAT) which are different from various statistical techniques.
•
To equip the students with skills to perform data analysis and conclusions independently with special focus on Data Mining (DM) applications with SAS Enterprise Miner. REFERENCE BOOKS
AUTHOR / PUBLICATION
Data Mining Techniques: For Marketing, Sales, and Customer Relationship Management
Berry and Linoff - Wiley Computer Publishing; 2 Ed, 2004.
Applied Multivariate Techniques
Sharma Subhash - John Wiley & Sons, 1996
Using Multivariate Statistics’
Tabachnick B.G. & Fidell L.S - Allyn & Bacon, 1996.
Multivariate Data
Hair J.F, Anderson R.E., Tatham R.L, Black W.C - Hair. Pearson Education, 2003.
Detailed Syllabus
examples of data screening – Outlier analysis Residual analysis – Generalized linear models – types of sum of squares type I II III IV – Concept of AIC & SBC - Data partition into Training validation and testing
Introduction to Data Mining Applications in Marketing and Customer Relationship Management: A Statistical perspective of Data Mining - Analytic Customer Relationship Management - Tasks performed with data mining - Virtuous Cycle of Data Mining Applications of data mining – Concept of Learning , Knowledge discovery, Analytical Intelligence, Enterprise Intelligence
Receiver operating characteristics (ROC) : Concept , Construction and inferences – Area under the curve (AUC) – Multiclass problems – Volume under the curve(VUC) Logistic regression: Concept of odds ratio
Examining data using SAS: Testing the assumptions of multivariate analysis- BLUE Assessing individual variables Vs variate Normality – Heteroscedasticity, autocorrelation & Multicollinearity – Identification and solutions -absence of correlated errors Important issues in data screening - Complete PG Program in Business Administration
- Wald’s confidence interval and construction – concept of Cordant discordant and tied pairs Basic concepts of logistic regression - Logistic regression with only one categorical variable Logistic regression and contingency table analysis - Logistic regression for combination of 1
Class of 2008
categorical and continuous independent variables – Stepwise backward and forward regression methods in multivariate logistic regression Comparison of logistic regression
Rules: Concept of support confidence lift and gain Defining Market Basket Analysis - Three Levels of Market Basket Data - Order Characteristic - Item Popularity - Tracking Marketing Inventories - Clustering Product Usage - Association Rules - Actionable Rules Trivial Rules - Inexplicable Rules - Building Association Rules - Choosing the Right Set Of Items - Product Hierarchies Help to Generalize Items - Virtual Items Go Beyond the Product Hierarchies - Data Quality Anonymous Versus Identified - Generating Rules From All the Data - Calculating Confidence - Calculating Lift The Negative Rule - The Problem of Big Data
and discriminant analysis Decision Trees: Introduction - Growing a decision tree - concept of logworth – algorithms chaid & cart importance of variable selection Test for choosing the best split – Pruning Extracting rules from trees - Alternate representations for decision trees - Decision trees in practice Artificial Neural Networks: History - Real Estate Appraisal – Concept of a link function Neural Networks for Directed data Mining Neural Net - The Unit of a Neural Network Feed-Forward Neural Networks - Back Propagation Heuristics for using Feed-Forward Back Propagation Networks - Choosing the Training Set - Coverage of Values for All Features - Number of Features - Size of Training Set - Number of Outputs - Preparing the Data - Features with Continuous Values Features with Ordered, Discrete (Integer) Values - Features with categorical values - Other Types of Features - Interpreting the results - Neural Networks for Time series - Example: Finding Clusters - Lessons Learned
Suggested Schedule of Sessions Topic
Time Series Forecasting: Stationarity non stationarity Exponential Smoothing - ARIMA Models - AR Process - Moving Average Process - ARMA Process - Box Jenkins Methodology – Time series construction of rare events SAS Programming: Basics of Programming Data input through programming - Data Steps and Proc steps Simple SAS Programming – Construction of charts and plots using SAS steps
No of Sessions
Introduction
1
Examing Data (Enterprise Guide)
3
Exercises on Examing Data
2
Logistic Regression (LR) (Enterprise Guide)
3
Exercises on LR
2
Decision Trees (DT) (Enterprise Miner)
3
Exercise on DT
2
Artificial Neural Networks (ANN) (Enterprise Miner)
4
Exercise on ANN
2
Time Series Forecasting (TSFS)
3
Exercise on TSFS
1
SAS Programming (BASE SAS)
3
Market basket Analysis (MBA)(Enterprise Guide)
3
Exercise on MBA
1
Total Sessions
33
Market Basket Analysis and Association PG Program in Business Administration
2
Class of 2008
PG Program in Business Administration
3
Class of 2008
