Applying Statistical Concepts using SAS®
Duration: 3 days
If you experience any problems with our on-line booking service, you can call us on
+44 (0)131 225 6952.
Description
This course is designed for predictive modellers and data analysts with basic
SAS programming experience. The issues and techniques discussed in this course are directed
toward database marketing, credit risk evaluation, fraud detection and other predictive modelling
applications from banking, financial services, direct marketing, insurance, and telecommunications.
Objectives
This course is an excellent follow-on to the SAS® Essentials: An Introduction to
SAS® Programming course for anyone using statistics in a business
environment. It covers a range of introductory statistical topics and uses SAS to carry out
analyses. Emphasis will be placed on the interpretation of the results.
After completing this course, you should be able to:
- see how statistics can be used to answer business problems
- interpret some simple summary statistics
- assess the precision of your statistics
- examine ways of testing business questions
- examine relationships between variables
- produce predictions of target variables
- explain why categorised data are treated differently
- know what to do if your target variable is binary (e.g. Yes/No, Default/Repay?)
Prerequisite Skills
Before attending this course, you should:
- have an understanding of basic mathematical concepts e.g. calculating percentages
- have some familiarity with basic statistical concepts such as variability may be beneficial
- have completed a SAS® Essentials: An Introduction to
SAS® Programming course
SAS® System Modules used
This course covers SAS/STAT® and touches on SAS/GRAPH®.
Delegates can benefit from this course even if SAS/GRAPH® is not installed at their
site.
Course Topics
Business Problems and Statistical Solutions
- Understanding the purpose of statistics
- Calculating some simple summary statistics
- Interpreting output from the MEANS and UNIVARIATE procedures
- Examining the variability of data, why can we never be sure?
Testing Business Questions
- Introduction to terminology for testing questions
- What is a t-test and when is it used?
- How can I calculate and interpret a p-value?
- Interpreting output from the UNIVARIATE and TTEST procedures
- What is analysis of variance and how is it different to a t-test?
- Performing analysis of variance with the GLM procedure
Categorising Data - Why is it different?
- Examining categorised data with the FREQ procedure
- Examining and testing for an association between two variables
- Calculating and interpreting the chi-square test for association
Is There a Relationship Between Two Continuous Variables?
- Exploring the relationship between two continuous variables
- Measuring a linear relationship using correlation
- Interpreting the output from the CORR procedure
- Understanding the misuses of correlation statistics
- Defining the linear relationship by using regression analysis
- Interpreting the output from the REG procedure
- What are the assumptions of regression analysis and how can we check them?
How can we look at more than two continuous variables?
- Is our target variable related to more than one variable?
- Can we get better predictions by using more variables?
- How can we select the 'best' variables?
What if our target variable is binary?
- Why do we need to do something different?
- What is logistic regression and how does it work?
- How can I interpret the results from the LOGISTIC procedure?
- What is an odds ratio and why is it useful?
Output Delivery System (optional)
- How the Output Delivery System can help
Reserve your place on this Newtyne SAS Training Course
|
