//indicate the path and working folder
cd "C:\Users\Aki\Documents\stata"

log using jan31

***********EXPORING DATA********************************************************
**view and edit raw data

**1. load data
sysuse auto, clear

**2. browse it
browse

browse make price

**3. describing and summarizing data
describe 
//gives an information about where the data is located, observation count, number of variables, size of database, var names and labels, storage type, etc.

summarize 
//gives summary stats for the variables (mean, st dev, min, max)

sum length trunk turn 
//summary stats for specific variables

**you can specify summarize with different conditions
summarize price if foreign==1
//will give sum stats for variable price if the car is produced abroad
summarize price if foreign==0

codebook foreign  //dummy (binary) variable
codebook price  //integer
codebook rep78 //most probably, categorical variable

summarize price if rep78<3
//summary stats for price of cars with repair record less than 3 (=1 or =2)
summarize price if rep78>=3
//summary stats for price of cars with repair record more than or equal to 3
summarize price if rep78!=3
//summary for price of cars with repair record not equal to 3 (=1, 2, 4, 5)

summarize price if foreign==1 & rep78>=3 
//summary for price of cars that are foreign and with rep record >=3 (match BOTH conditions)

summarize price if foreign==1 | rep78>=3
//summary for price of cars that are either foreign or with rep record >=3 (match ANY of these conditions)

sum foreign
//not extremely useful for binary and categorical variables

***TABULATING AND CREATING TABLES******
**for categorical variables, tabulating is very useful
tab foreign
tab foreign, nolabel
//to see what variable corresponds to what label

tab rep78

*two way tabulation
tab foreign rep78
tab foreign rep78, col row

******MISSING VALUES************************
***Stata reads missing values as positive infinity
misstable summarize

*Why need to be careful with missing values?
tab rep78
tab rep78, miss

sum price if rep78>=5
sum price if rep78>=5 & rep78!=.

sum price if rep78!=.
sum price if rep78==.

**if want to recode missing values, mvencode/mvdecode
mvencode *, mv(-99)
mvdecode *, mv(-99)

translate jan31.smcl jan31.log
log close