|
Apr 15, 2013 |
Ms. Kelley Mizukami |
Surviving Survival Analysis
Chocolate Cake Seminar Series on Statistical Applications
Apr 15, 2013
FO 5-Room 116
Let’s pretend you participate in a weight loss study. The study isn’t over yet, but you got a very luring job offer that requires you to move out of state and you cannot continue in this study. Should your data be thrown out and not used for the analysis after all since you never completed the study? The answer is “No, not at all”. The talk will explain how to model times to events (e.g., 20% excess body weight loss) in presence of drop-outs. Two survival analysis techniques, Kaplan-Meier curve and Cox model, will be studied through SAS and SPSS applications.
view
|
 Presentation Slides Sample Data |
|
|
Mar 18, 2013 |
Dr. Olga Korosteleva |
Non-poisonous Poisson Regression at Your Fingertips
Chocolate Cake Seminar Series on Statistical Applications
Mar 18, 2013
FO 5-Room 116
Have you ever counted how many cars are in front of you at McDonald’s drive-thru? Is it a number between zero and infinity? Well, just notice the hour of the day and you got yourself a Poisson regression. Attend the talk and learn how to model properly the relation between the number of cars and the hour of the day. You won’t be bored. We promise. Applications in SAS and SPSS will definitely keep you awake.
view
|
Presentation Slides Sample Data and Code |
|
|
Mar 4, 2013 |
Mr. David Ramirez |
Be an Explorer with Exploratory Data Analysis!
Chocolate Cake Seminar Series on Statistical Applications
Mar 4, 2013
FO 5-Room 116
Got data? Now what? Come to the talk and find out what your first exploratory steps might be. Examples in statistical packages SAS and SPSS will be considered. There will be time for questions and discussion at the end of the talk.
view
|
Presentation Slides Sample Data |
|
|
Feb 18, 2013 |
Mr. William Guo |
Get Empowered by Power Analysis!
Chocolate Cake Seminar Series on Statistical Applications
Feb 18, 2013
FO 5-Room 116
How can you tell how many people you need for your study? Is it ten, twenty, sixty, one hundred? The number of participants should not be determined by time and budget alone. Statistical rigor, called power analysis, must be applied. The talk will expose the details with applications in SAS and other specialized softwares.
view
|
Presentation Slides |
|
|
Nov 19, 2012 |
Dr. Olga Korosteleva |
Multinomial Logistic Regression Models, Your Turn!
Chocolate Cake Seminar Series an Statistical Applications
Nov 19, 2012
FO5-Room 116
If your outcome variable assumes not just two, but several values, then the binary logistic regression is out of question. What models can you use then? Come to the talk and find out. SAS and SPSS examples will be presented. There will be time for questions and discussion at the end.
view
|
Presentation Slides Sample data and code |
|
|
Nov 5, 2012 |
Dr. Olga Korosteleva |
Binary Logistic and Probit Models at Work
Chocolate Cake Seminar Series an Statistical Applications
Nov 5, 2012
FO5-Room 116
If your outcome variable assumes not just two but several values, then the binary logistic regression is out of question. What models can you use then? Come to the talk and find out. SAS and SPSS examples will be presented. There will be time for questions and discussion at the end.
view
|
Presentation Slides Sample data and code |
|
|
Oct 22, 2012 |
Ms. Kelley Mizukami |
What are One-Way and Two-Way Analyses of Variance?
Seminar Series: Statistical Applications
Oct 22, 2012
Have you ever wanted to test if the means of several groups were the same or not? Then analysis of variance (ANOVA) is what you need. This presentation will discuss how to test whether several means are the same in designs with one factor as well as two factors using such statistical software packages as SAS and SPSS. There will be time for questions and discussion at the end.
view
|
 Presentation Slides Sample data and code |
|
|
Oct 1, 2012 |
Dr. Olga Korosteleva |
Having Fun with Linear Regression Models
Seminar Series: Statistical Applications
Oct 1, 2012
FO5-Room 116
If you would like to study the association between a continuous outcome variable and several independent variables, a linear regression is your tool. The talk will display how to develop well-fit models using such statistical software packages as SAS and SPSS. There will be time for questions and discussion at the end of the presentation.
view
|
Presentation Slides Sample data and code |
|
|
May 2, 2012 |
Dr. Olga Korosteleva |
Binary and Multinomial Logistic Regressions at Work
RIMI Brown Bag Seminar Series on Statistical Applications
May 2, 2012
What if the variable you are about to model assumes only a handful of different values? Can you say it is a continuous variable and use a linear regression model? The answer is enthusiastically, “No!” The talk will explain why it is so and introduce you to appropriate regression approaches. SAS and SPSS statistical software packages will be employed. Basic knowledge of linear regression modeling will be required. There will be time for questions and discussion after the presentation.
view
|
|
|
