Exploring Generalized Regression Models in Functional Data Analysis
Dive into the world of Generalized Regression models with Genreg in JMP Pro, a versatile platform for handling various response types like skewed, censored, and non-numeric data. Learn how Genreg can offer flexibility in model building beyond traditional linear regression, accommodating unique response scenarios effectively.
- Generalized Regression
- Functional Data Analysis
- Predictive Modeling
- Response Types
- Variable Selection
Download Presentation
Please find below an Image/Link to download the presentation.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. Download presentation by click this link. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.
E N D
Presentation Transcript
Breakout Room #3 Breakout Room #3 Functional Data Analysis Functional Data Analysis Predictive Modeling Un-session Timing (CET) 12:30-12:45 rooms 12:45-13:25 Welcome, get to breakout Breakout room discussions 13:25-13:35 rooms 13:35-14:15 Break, change breakout Turn on Turn on your video your video Raise your hand Raise your hand to start a conversation to start a conversation Breakout room discussions 14:15-14:25 rooms 14:25-15:05 Break, change breakout Breakout room discussions 15:05-15:10 Passing to session main hall 15:10-15:30 Recap sessions & Wrap Up Tom Donnelly Tom Donnelly Principal Systems Engineer Principal Systems Engineer Florence Kussener Florence Kussener Sr Systems Engineer Sr Systems Engineer Use the chat to comment on the discussion Use the chat to comment on the discussion
Turn on your video Turn on your video Raise your hand to start a conversation Raise your hand to start a conversation Use chat to comment on the discussion Use chat to comment on the discussion What is What is Genreg Genreg The Generalized Regression platform in JMP Pro can be your go-to place for building regression models. From the Fit Model dialog, switch the personality to Generalized Regression.
Turn on your video Turn on your video Raise your hand to start a conversation Raise your hand to start a conversation Use chat to comment on the discussion Use chat to comment on the discussion Our Our r response esponse isn t isn t always normal always normal Generalized Linear Models Usually when we build regression models, we think about the response being normally distributed. But that isn t always the case 1. Skewed responses like income (Gamma, lognormal, ) 2. Outliers (Cauchy or Quantile regression) 3. Non-numeric responses like red/blue/green (Nominal/Ordinal Logistic) 4. Censored, where we only have partial information (Ex: ??> 40) Genreg can handle all these situations and more.
Turn on your video Turn on your video Raise your hand to start a conversation Raise your hand to start a conversation Use chat to comment on the discussion Use chat to comment on the discussion A Variety of Response Types A Variety of Response Types All in one place
Turn on your video Turn on your video Raise your hand to start a conversation Raise your hand to start a conversation Use chat to comment on the discussion Use chat to comment on the discussion Variable Selection Variable Selection It s unusual (impossible?) to know what factors to include in our model. Including unnecessary effects can lead to bad decisions and bad predictions. Genreg includes a variety of automated variable selection methods that can help us know which factors to include in our model.
Turn on your video Turn on your video Raise your hand to start a conversation Raise your hand to start a conversation Use chat to comment on the discussion Use chat to comment on the discussion Variable Selection Variable Selection A variety of methods Genreg provides a handful of selection methods for any predictor type, even when you have collinearity. Step based methods are old-school algorithmic approaches. Penalized regression methods are more modern and do estimation and selection simultaneously.
Turn on your video Turn on your video Raise your hand to start a conversation Raise your hand to start a conversation Use chat to comment on the discussion Use chat to comment on the discussion So what is So what is Genreg Genreg? ? Genreg is your one-stop shop for building regression models in JMP Pro. regardless of what kind of response/predictors you have whether you have observational data or a designed experiment.
Turn on your video Turn on your video Raise your hand to start a conversation Raise your hand to start a conversation Use chat to comment on the discussion Use chat to comment on the discussion And it s interactive And it s interactive
Breakout Room #4 Breakout Room #4 Generalized Regression Generalized Regression Resources Resources Mastering JMP Mastering JMP jmp.com > Learn JMP jmp.com > Learn JMP Application area > Statistics, Predictive Modeling and Data Mining Application area > Statistics, Predictive Modeling and Data Mining Clay Barker Clay Barker Principal Research Statistician Developer Principal Research Statistician Developer JMP Learning Library JMP Learning Library jmp.com > Learn JMP jmp.com > Learn JMP Data Mining and Predictive Modeling Data Mining and Predictive Modeling Multivariate Methods Multivariate Methods Hadley Meyers Hadley Meyers Sr. Systems Engineer Sr. Systems Engineer User Community User Community community.jmp.com community.jmp.com Learn JMP Learn JMP Turn on your video Turn on your video Raise your hand to start a conversation Raise your hand to start a conversation Use chat to comment on the discussion Use chat to comment on the discussion Statistical Thinking and Industrial Problem Solving (STIPS) Statistical Thinking and Industrial Problem Solving (STIPS) https://www.jmp.com/en_us/online https://www.jmp.com/en_us/online- -statistics Correlation and Regression Module Correlation and Regression Module Predictive Modeling and Text Mining Module Predictive Modeling and Text Mining Module statistics- -course.html course.html