R-bios6643-L09sas.Rmd

---
title: "BIOS6643. L09 Specifying both G and R"
cheauthor: "EJC"
-- date: "" 
header-includes:
   - \usepackage{amsmath}
   - \usepackage{float}
output: pdf_document
---

\newcommand{\bi}{\begin{itemize}}
\newcommand{\ei}{\end{itemize}}
\newcommand{\itt}{\item}


```{r setup, include=FALSE, echo=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

<!-- -->


```{r installp, echo = FALSE, eval=T, include=F}
library(SASmarkdown)

```

```{r include=FALSE}
# set up the options so that knit knows where you SAS executable is
# set the line size to be easily readable on letter size paper, portrait
# and set the knitr options using opts_chunk$set().
sasopts <- "-nosplash -linesize 75"
saspath <- "C:/Program Files/SASHome/SASFoundation/9.4/sas.exe"
## those two parts should have been universal 
## just copy and paste into your Rmarkdown files
knitr::opts_chunk$set(engine = "sas", 
                      engine.path = saspath, 
                      engine.opts = sasopts, 
                      comment = NA)

```


The RANDOM statement may be used with TYPE=UN
  and possibly the GROUP= option to specify the among-individual
  covariance matrix G (possibly different by levels of GROUP).  

The REPEATED statement may be used with various TYPE= specifications,
   and possibly the GROUP = option with SUBJECT = to 
  specify the within-individual covariance matrix R_i.
  

```{r engine="sas", engine.path="C:/Program Files/SASHome/SASFoundation/9.4/sas.exe", engine.opts="-nosplash -linesize 75", collectcode=TRUE}


proc import DATAFILE='C:\Users\juarezce\OneDrive - The University of Colorado Denver\BIOS6643\BIOS6643_Notes\data\dental.csv'
  replace out=dental0 dbms=csv; run;

data dental; 
  set dental0;
  time=age;
run;

 proc print data=dental(obs=2);run;

  
  
  title '(a) DIAGONAL WITHIN-CHILD COVARIANCE MATRIX R_i';
title2 'WITH CONSTANT VARIANCE SAME FOR EACH GENDER';
title3 'SAME G MATRIX FOR BOTH GENDERS';
proc mixed method=ml data=dental;
  class gender id;
  model distance = gender gender*age / noint solution;
  random intercept age / type=un subject=id g gcorr v vcorr;

run;
  
  
/******************************************************************
  
   MODEL (b)

    Fit the same model as (a) but with a separate diagonal Ri matrix for
   each gender.  Thus, there are 2 separate variances sigma^2_(G and B)
   specified using GROUP=GENDER in the REPEATED statement

  
*******************************************************************/
    
title '(b) DIAGONAL WITHIN-CHILD COVARIANCE MATRIX R_i';
title2 'WITH SEPARATE CONSTANT VARIANCE FOR EACH GENDER';
title3 'SAME G MATRIX FOR BOTH GENDERS';
proc mixed  method=ml data=dental; 
  class  id gender;
  model distance = gender gender*age / noint solution;
  repeated / group=gender subject=id r=1,4 rcorr=1,4;
  random intercept age / type=un subject=id g gcorr v=1,4 vcorr=1,4;
run;




  /* */
  
```