* Clear existing data and working space 
clear all

* Open data from bigblue server 
* use c:\eco304\mls2017chi.dta
use https://bigblue.depaul.edu/jlee141/econdata/eco304/mls2017chi.dta

* Log file to save output
log using lab1output, replace 
log close
type lab1output.smcl

* Describe and list data
desc
list in 1/6 
list houseid hprice sqft zip
list houseid hprice if zip == 60618 

* Tabluate for frequencies 
tab zip 
tab zip fireplace 
tab zip bedroom  

* Create new variables and labeling 
gen lhprice = log(hprice)
gen sqyard  = sqft/9
gen lsqft   = log(sqft)
gen age_sq  = agebld*agebld 

label var hprice   "House price in $"
label var lhprice  "Log of house price"
label var sqyard   "Square Yard"
label var lsqft    "Log of Square Feet"
label var agebld   "Age of House"
label var age_sq   "Square of Age"

* Create dummy variables using the existing variables
gen oldhouse= agebld > 65 
tab oldhouse 
tab zip, gen(z)


* Descriptive Statistics 
sum
sum, detail
sum hprice 

* Covariance and Correlation
correlate hprice sqft , covariance 
correlate hprice sqft

correlate hprice sqft if zip == 60002, covariance  
correlate hprice sqft if zip == 60002

* Confidence Interval and Hypothesis test
ci mean hprice sqft  
ci mean hprice sqft, level(99)

ttest hprice == 300000 

* Charts 
hist hprice
hist hprice, bin(16)
hist hprice, bin(12) frequency
hist hprice, bin(12) percent
hist hprice if zip == 60618, bin(12) percent

twoway (scatter hprice sqft)
twoway (scatter hprice sqft) if hprice > 200000 
twoway (scatter hprice sqft) if hprice < 200000 

* Saving Data
save c:\eco304\my_hprice