This repository contains SAS related Projects / Models / Codes etc .

#Modular Code File -

Run transformation code befor analysis code Steps -

1. Read in Data
2. Keep Only Native Variables Needed
3. Apply Exclusions
4. Add transformed variable from Data Dictionary
5. .... & last step would be Analysis Code

#Part 1 : Descriptive Analysis

Code with prefix 100s - trnsformation Code 200s - Plot Code 300s - Descriptive analysis 400s - Extras

#part 2 -

Regression Analysis

500 - Linear 600 - Logistic 700- extra

#Dataset namingconvension - BRFSS_ - Native dataset Copy of BRFSS_a to BRFSS_b ( Remove unnecessary Columns) copy BRFSS_b top BRFSS_c and remove exclusions

#Hypothesis Declaration - As we are using two regeression for analysis - Linear and logistic so we must schoose a hypothesis which will be best for descriptive analysis and regression .

#Component of Hypotesis - Subpopulation ( Sample Populaton )

We need two hypothesis as output from Linear regression is continuous and that from logistic regression is binary

=========================

Subpopulation - Veterans Exposure - Having Diabetes Disease - Linear regression ( Avg sleep per night) logistic regression ( Having Asthma)

1. Among veterans diabetes status is statistically significantly associated with average hours of sleep per night

2. Among veterans diabetes status is statistically significantly associated asthama status

In SAS the general sas file format is *.sasbdat but this file is bigger (in size) in nature compare to the .csv from which it as generated or the .csv having the same data. * For saving too huge data SAS came with up a new format called .xpt.

.xpt format similar to zip format of .sas7bdat , we need to unpack .xpt to .sas7bdat file

We also need special lib name for for .xpt #Some code are below for reference

Typical libname :

Libname p "c:\sas"; run;

for reading *.xpt file

Libname p "c:\sas"; run;

libname XPTfile xport "c:\sas\name.xpt"; run;