cd "C:\Users\Aki\Documents\stata"
log using feb14.log

****DATA MANIPULATION. PART 2******

**naming and labeling variables****
*change a name of a variable
sysuse auto
rename rep78 repair_record
rename repair_record rep78

**Stata differentiates between upper and lower cases
sum Price   //no such variable
sum price

**renaming many variables at once
rename * *1978
rename *1978 *

rename * *a
rename *a *

rename m* M*
rename M* m*

**labeling a variable
label variable rep78 "repair record of cars in 1978" 
tab rep78

**labeling values (useful for dummy and categorical variables)
label dir
tab foreign

**step 1. define a label
label define repair 1 "A" 2 "B" 3 "C" 4 "D" 5 "E"
label dir

**step 2. connect a label to a variable
label values rep78 repair
tab rep78


***if it was aboue labeling ethnicities
*** label define ethnic 1 "kyrgyz" 2 "uzbek" 3 "russian" 4 "other"
***label values ethnicity ethnic 

*****EGEN OR EXTENDED GENERATE******
help egen

***you want to generate a new variable, ten equal deciles for price
egen deciles = cut (price), group (10)
tab deciles

**we want to have average value of price, miles per gallon, weight
egen rowmean = rowmean (price mpg weight)
browse price mpg weight rowmean

tab rep78, sum (price)
egen pricerepair = mean (price), by(rep78)
tab pricerepair

****if you want to generate an average cons per household
**hh1  cons1
**hh1  cons2
**hh1  cons3
**hh2  cons2
**hh3  cons1
**hh3  cons3
**egen meancons = mean(cons), by(hhid)

*** egen totalincome = total(income), by (hhid)

****DUMMY VARIABLES
**dummy variable is a binary variable: two categories, usually 0 and 1
*e.g. gender (female and male), location (urban/rural), vote (yes/no)

**gen a dummy variable for price: lowprice and highprice
gen highprice = 1 if price>6165
replace highprice = 0 if price<=6165
//replace highprice = . if price==.   in case of any missing observations in the original variable price
tab highprice
**be careful with missing observations!

**gen a dummy var from a categorical variable
tab rep78

gen A = 1
replace A = 0 if rep78!=1

gen B = 1
replace B = 0 if rep78!=2

gen C = 1
replace C = 0 if rep78!=3

///continue with all the categories
//2 lines of codes for each category

**a shortcut version of doing this:
tab rep78, gen(dummy)
//5 new dummy variables are created. need to rename them
rename dummy1 A
rename dummy2 B

***KEEEPING AND DROPPING VARIABLES****
**to remove a variable from a dataset
drop dummy5
drop A B 

**keep only selected variables****
keep make price foreign trunk

**keep variables that have "e" in their name
keep *e*

**drop variables that are not domestic
tab foreign
drop if foreign==1

log close