* this exercise uses the "census" data set available via https://www.openintro.org/stat/extras.php

* clear the data
clear

* describe the data (variable name, storage type, etc.)
des

* summarize the data (variable, observations, mean, standard deviation, min, max)
sum

* look at data in read-only format
browse

* look at data with the option to make changes
edit

* drop variables
drop censusyear

* rename variables
rename totalfamilyincome finc
rename totalpersonalincome pinc
rename racegeneral race
rename maritalstatus marital
rename statefipscode state

* destring, replacing "NA" with .
destring finc, replace ignore("NA")

* mean of a variable, including confidence interval around mean 
mean pinc

* generate a new binary variable based on existing string variable
gen male = 1 if sex == "Male"
replace male = 0 if male == .

* histogram
hist pinc

* box and whiskers graph
graph box pinc

* pie charts
gen one = 1
graph pie one, over(sex)
graph pie one, over(race)

* scatter plot
scatter pinc age

* scatter plot with linear fit
scatter pincr age || lfit pincr age

* scatter plot with quadratic fit
scatter pincr age || qfit pincr age

* regression of personal income on age
regress pinc age

* regression of personal income on age and age squared
gen age2 = age^2
reg pinc age age2

* regression of personal income on age, age squared, and gender
reg pinc age age2 male