Notes on STATA
Some nitty-gritties that economists remember but I don't
By Harshvardhan in STATA economics
February 16, 2022
Conventions through this page
x
is the name of the variable.
Set Working Directory
cd "/directory/"
Use a Dataset
use data.dta, clear
Clear all items from memory
clear all
See documentation for details.
Display all items
list
Display first element only (useful for a scalar)
di x
Setting number of observations
qui set obs 30
qui
says this command is executed quietly, i.e. with no output.
Summary of a variable
summarize x
This will number of observations, mean, standard deviation, minimum and maximum.
Describe
describe x
This will tell us the type of variable it is.
Functions
Functions in Stata are called Programs. See documentation.
Below is a sample program named onesample
that generates 30 uniform random numbers and calculates their mean. We do need to specify rclass
to specify the kind of return for the function. Other options is eclass
.
drop _all
is different from drop all
; the former drops all observations and the latter looks for a variable named all
and deletes that.
program onesample, rclass
drop _all
qui set obs 30
gen x = runiform()
summ x
return scalar mean = r(mean)
end
Monte Carlo Simulation
If the function is defined as above, we can use the following codes to perform Monte Carlo simulation.
* Simulate it 10000 times
simulate xbar = r(mean), seed(0) reps(10000): onesample
Visualisation
Histogram
hist x, width(0.1) title("Histogram of x")
See documentation for details.