* Example 4.2.1 

* Imort data set to Stata memory
import delimited using "data/EXA_C04_S02_01.csv", clear

* Take a look at data
describe
list

* Rename the variable to a shorter name
rename numberofprograms nprg

* Take a look at data, again
describe
list

* Calculate relative percentage and cummulative percentage
tabulate nprg [fweight = frequency]

* Plot probability density function
histogram nprg [fweight = frequency], discrete color(blue) fintensity(inten10) barwidth(0.8) xlabel(#8)


* Generate new variables for relative frequency
egen totalFreq = sum(frequency)
gen Frac = frequency/totalFreq
label variable Frac "Relative Frequency"
drop totalFreq 
* Take a look at data, again
list

* Generate cummulative distribution
cumul nprg [fweight = frequency], generate(cum)
* Take a look at data, again
list

* Calculate mean, variance and standard deviation
summarize nprg [fweight = frequency], detail

* Make the plot of cumulative distribution
gen cum1 = cum[_n - 1]
replace cum1=0 if cum1==.
rename cum cum2
reshape long cum, i(nprg) j(idx)
label variable cum "Probability"
set obs 17
replace nprg = 0 if nprg==.
replace cum = 0 if cum==.
set obs 18
replace nprg = 9 if nprg==.
replace cum = 1 if cum==.
sort nprg idx
line cum nprg, sort xlabel(#10)