******************************
** Application 2 *************
******************************
** Mar 26, 2021 **************
******************************

** change directory **
	cd "E:\My Drive\Heal STATA Trainings\Training 2\data"

** import data **
	import excel using Titanic.xlsx, first sh("data") clear
	
** question 1  - number of passengers **
	count 
	// this will count the total number of observations in the data set, but passengers might have missing values
	// so you could double check with the following command, and look at the number of observations
	sum PassengerId

** question 2 - how many were male **
	count if Sex=="male"
	tab Sex

** question 3 - how many passengers were male and >25 **
	tab Sex if Age>25
	sum PassengerId if Sex == "male" & Age > 25
	
** question 4  - average fare female passenger **
	sum Fare if Sex=="female" 
	
** question 5 - average for for queenstown OR southampton **
	sum Fare if Embarked == "Q" | Embarked == "S"
	
** question 6 - create dummy variable **
	gen female=0
	replace female=1 if Sex=="female"
	
	//or you could do the following if you wanted to create one for male
	
	gen male=0
	replace male=1 if Sex=="male"
	
	//because female is a binary variable, if you create that one, you do not need to create another binary for male
	
** question 7 group variables **
	// first create three variables with a value of 0
	gen class1=0
	gen class2=0
	gen class3=0
	
	//then replace them to equal one if the condition is met
	replace class1=1 if Pclass==1
	replace class2=1 if Pclass==2
	replace class3=1 if Pclass==3
	
** question 8 - crosstab check **
	// whenever you create dummy variables, it is always a good idea to check
	tab Pclass class1
	tab Pclass class2 
	tab Pclass class3 
	
	// the way that this is a check is that if Pclass=1 then class1 should be =1.
	
** question 9 - city variables **
	gen city = "Queenstown"
	replace city = "Southhampton" if Embarked == "S"
	replace city = "Cherbourg" if Embarked == "C"

	
** question 10 - who survives **
	//percentage of males who survived
	sum Survived if female==0
	
	//percentat of female who survived
	sum Survived if female==1
	
	//Now conduct the ttest, which will give you the same answers as the code above
	ttest Survived, by(female)

** question 11 - survival rate by age **
	//first generate variable for old
	gen old=0
	replace old=1 if Age>35
	
	//check accuracy of variable
	tab Age old
	
	// notice that old equals one only after the individual is older than 40
	// you will see that the youngest age is 0.42, is that an error?
	// not if the youngest was a baby that was only 5 months old (.42 of a year)
	
	ttest Survived, by(old)