- General Linear Modeling -Outline Part 1 Outcome variable is considered continuous
- General Linear Modeling -Outline Part 3 Examining moderators (i.e interactions)
- This computer lab describes how to do the following in SAS/R/STATA. 1. Read comma delimited data (.csv)
- Binary Outcomes -Generalized Linear Models Part 1 2 by 2 tables
- Count outcomes -Poisson regression Generalized Linear Models Part 3
- STATA data analysis of the living will study in homeless persons USING the logistic function (and the character variable trtgroup)
- Examining Predictability and Goodness of Fit for logistic regression proc logistic data = birthweight descending;
- Ordered logistic regression in SAS/STATA and R Example looking at Women's baseline weight status (1,2,3,4) at time of pregnancy
- Ship damage example Poisson regression in SAS ## Ship damage data from McCullagh and Nelder (1989), Sec 6.3.2. Each row
- Automated Variable Selection Methods Variables selection technique -chooses which variables make the model "fit" better
- VARIABLE SELECTION EXAMPLE in SAS Run the following code to generate a dataset A with 1000 observations and 11 variables x1-x10
- PUBH 7402 Biostatistics Modeling and Methods This course is about modeling data and using the estimates of those models
- Taken directly from http://www.ats.ucla.edu/stat/sas/webbooks/reg/chapter2/sasreg2.htm
- Examining linearity The continuous predictors in the model are assumed to be linearly related to the outcome. In our
- Estimating different slopes resulting from interactions (response to Homework 5)
- Logistic regression in SAS and R How is a mother's gestational weight gain related to the the probability of the baby being born with
- Examining Goodness of Fit for logistic regression proc logistic data = birthweight descending;
- J Clin Epidemiol Vol. 51, No. 10, pp. 809816, 1998 0895-4356/98/$-see front matter Copyright 1998 Elsevier Science Inc. PII S0895-4356(98)00066-3
- Ordered logistic regression in SAS and R By default if the outcome variable has more than 2 levels, proc logistic will perform an ordered
- Ship damage example Poisson regression in SAS ## Ship damage data from McCullagh and Nelder (1989), Sec 6.3.2. Each row
- Automated Model Selection -Stepwise procedures Variables selection technique -chooses which variables make the model "fit" better
- R reshape command demonstration Solberg 1 R Directions for the Reshape Command
- Reexamining the alcohol abstaining data accounting for potential correlation due to clustering of individuals within clinics.
- Estimating survival curves and differences between curves for different groups (Kaplan Meier, Log-rank and Wilcoxon) in SAS and R
- Partial support for this work was provided by the National Science Foundation's Course, Curriculum, and
- Repeated Measures Analysis -Correlated Data Analysis -Multilevel data analysis -Clustered data -Hierarchical
- Longitudinal data analysis Pre-post treatment studies
- Paper 205-30 Using the Proportional Odds Model for Health-Related Outcomes: Why,
- Linear Mixed Models Appendix to An R and S-PLUS Companion to Applied Regression
- GEE compared to Mixed modeling for LINEAR LINK Pancreatic Enzyme example revisited
- General Linear Modeling -Part 2 Multiple regression -multiple predictors -uses and interpretations
- Using the method of Firth to obtain parameter estimates in the case of separation in logistic regression.
- Categorical outcome variables (Beyond 0/1 data) Generalized Linear Models Part 2
- Reexamining the alcohol abstaining data accounting for potential correlation due to clustering of individuals within clinics.
- Paper SA01_05 Code to Select the Best Multiple Linear Regression Model
- Generalized linear model From Wikipedia, the free encyclopedia
- Multiple (more than 2) time points -Longitudinal modeling Consider the following (simulated data) where a baseline measurement of Y was taken and then 2 followup
- Repeated Measures Analysis -Correlated Data Analysis -Multilevel data analysis -Clustered data -Hierarchical
- Numbers associated with the test for mediation in the GPA, MVPA, SPORTS data
- Pre-post analysis of homeless study problems of daily living data
- Using the EFFECTS package in R to get adjusted means and plot results Previously I demonstrated how to get the adjusted means from multiple regression including multiple
- Results from fitting SKIN CANCER DATA with Poisson regression including AGE in different ways
- PUBH 7402 Biostatistics Modeling and Methods Topics covered
- q GLMs for count data q More Examples. . .
- Estimating the Survival function (Kaplan-Meier) and testing with the log rank test
- In this computer lab we will see how to do the following in BOTH SAS and R. 1. Reading comma delimited data (.csv)
- General Linear Modeling -Outline Part 2 Multiple regression -multiple predictors -uses and interpretations
- Modeling Assumptions: The first 2 assumptions apply to any generalized linear model, the 3rd
- Generalized Linear Modeling -Logistic Regression Binary outcomes
- Adapted from www.graphpad.com Testing for equivalence is not just the opposite of testing for significant differences.
- General Linear Modeling -Part 1 Simple linear regression -1 continuous outcome with 1 continuous predictor
- Multinomial logistic regression proc logistic data = b descending;
- Statistical Computing Seminar Introduction to Multilevel Modeling Using SAS
- Examining and Examining Interactions (in SAS and R) First we examine whether the effect MVPA has on GPA is moderated by
- Heteroscedasticity -Non constant error variances Recall the general linear model
- Survival Data: Time from One Event to Another Initiating event Terminating event
- PROC LOGISTIC: Traps for the unwary Peter L. Flom, Independent statistical consultant, New York, NY
- Using STATA for mixed-effects models (i.e. hierarchical linear model) The XTMIXED function is for Multilevel mixed-effects linear regressions
- Frederick K.H. Phoa Statistics 120B Discussion Notes #3 (Spring 2007)