* Change directories and open a log file
cd "C:\Desktop\PBC_Analysis"
log using pbclog, replace

* Read in the data set
use pbc.dta, clear

* Explore the data set
describe
codebook

browse
list
list id drug sex
list in 1/10
list if sex == 1
list id drug if sex == 1

* Create new continuous variables
generate ageyrs = age/365.25
generate fuyrs = fudays/365.25

* Create new categorical variable
generate censored = .
replace censored = 1 if status == 0 | status == 1
replace censored = 2 if status == 2

* Explore the revised data set
describe

* Add labels to new variables
label variable ageyrs "Age (yrs)"
label variable fuyrs "Follow-up (yrs)"
label variable censored "2-level survival status"

* Overwrite some existing labels
label variable age "Age (days)"
label variable fudays "Follow-up (days)"
label variable status "3-level survival status"

* Add value labels
label define statuslabel 0 "Censored" 1 "Censored due to liver treatment" 2 "Dead"
label value status statuslabel

label define censoredlabel 1 "Censored" 2 "Dead"
label value censored censoredlabel

label define druglabel 1 "D-penicillamine" 2 "Placebo"
label value drug druglabel

label define sexlabel 1 "Female" 2 "Male"
label value sex sexlabel

label define asciteslabel 0 "No" 1 "Yes"
label value ascites asciteslabel

label define stagelabel 1 "1" 2 "2" 3 "3" 4 "4"
label value stage stagelabel

* Explore the revised data set
codebook
list in 1/10

* Summarize continuous variables, just the no. non-missing obs, mean, SD, min, and max
summarize fudays fuyrs age ageyrs bili chol album
* detail = percentiles (1, 5, 10, 25, 40, 75, 90, 95, 99), 4 smallest, 4 largest, n obs, mean, SD, variance, skewness, kurtosis
summarize fudays fuyrs age ageyrs bili chol album, detail
summarize ageyrs, meanonly
summarize ageyrs if sex == 1
summarize ageyrs if sex == 2

tabstat ageyrs, by(sex) statistics(count mean sd p25 median p75) missing
tabstat ageyrs, by(drug) statistics(count mean sd p25 median p75) missing

* Summarize categorical variables
tabulate sex
tabulate drug, missing
tabulate sex drug, missing column
* sex in rows, drug in columns

table stage sex drug, missing
* stage in rows, sex in (nested) columns, drug in "super"-columns

table stage sex drug, contents(freq mean ageyrs) missing

* Close the log file
log close