* slide 13:  reading in data
clear
cd "I:\Classes\StatComputingI"
insheet using "SCBC2004.csv", comma
clear
use SCBC2004.dta

/* slide 14:  exploring our dataset
 use d or describe */
d
d ercat
codebook 
codebook dodyr
sum
sum ercat
codebook ercat

* slide 16:  more exploration
tab race
table race
list race age if age<30
sort age

* slide 20
outsheet race age caseid prcat using "smalldataset.csv", comma replace

* slide 22
drop if race==1
keep ercat prcat stagen age grade

* read back in original data
clear
use SCBC2004.dta

* slide 23
by race, sort: sum age
bysort ercat prcat: sum age

tab ercat if stagen>1
tab ercat if graden~=.

bysort ercat prcat: sum age if ercat<9 & prcat<9

* slide 25
gen highgrade = 1 if graden>2 | graden<100
replace highgrade = 0 if graden<3

* cannot create new variable with same name: must drop it first
drop highgrade
gen highgrade=cond(graden>2,1,0)
replace highgrade = . if graden==.

* slide 26
drop highgrade
egen highgrade=cut(graden), at(-1,3,5)

drop highgrade
egen highgrade=cut(graden), at(-1,3,5) icodes

* slide 27
gen y = log(age)
gen x2 = age^2

gen z1 = runiform()
* uniform(2,4)
gen z2 =2+2*runiform() 

gen id= _n
bysort county: gen countyid=_n
 
drop z1 countyid y x2 id


* slide 28
* step 1:
gen z1=runiform()

* step 2
sort county z1
by county : gen countyid=_n
* step 3:  keep only 10 women in county
drop if countyid>10

* slide 29
* read in data with unformatted dates
clear
use Ohiosmall.dta

tab year
gen datedx=date(date_of_dx, "MDY")
format datedx %td
gen yeardiag=year(datedx)
list  date_of_dx dxdt_text2 dxdtchar datedx yeardiag if yeardiag==.

* slide 30
* reshape
clear
insheet using "I:\MUSC Oncology\Shirai, Keisuke\October2010\ceramide.csv"
reshape wide collecteddate - frombaselines1p, i(patient) j(cycle)

reshape long

clear
insheet using "ceramide2.csv"
rename cycle1totalceramidelevels totalceramidelevels1
rename cycle1diseasestatus diseasestatus1
rename cycle1c18ceramide c18ceramide1
rename cycle3totalceramidelevels totalceramidelevels3
rename cycle3diseasestatus diseasestatus3
rename cycle3c18ceramide c18ceramide3
rename cycle5totalceramidelevels totalceramidelevels5
rename cycle5diseasestatus diseasestatus5
rename cycle5c18ceramide c18ceramide5
rename cycle3daysfromstart daysfromstart3
rename cycle5daysfromstart daysfromstart5

reshape long daysfromstart diseasestatus totalceramidelevels c18ceramide , i(patient) j(cycle)
drop if totalcerami==.
replace daysfromstart=0 if cycle==1