home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Fritz: All Fritz
/
All Fritz.zip
/
All Fritz
/
FILES
/
BUSINESS
/
STATA3.LZH
/
DATAKIT.TUT
< prev
next >
Wrap
Text File
|
1988-08-29
|
8KB
|
248 lines
set output error
set display page 23
set more 1
#delimit ;
di _n(13) in wh
" ___ ____ ____ ____ ____ tm" _n
" /__ / ____/ / ____/" _n
"___/ / /___/ / /___/ Data.Kit: Programmed Extensions to Stata" _n
"-------------------------------------------------------------------------"
_n(2) ;
di in gr
"Data.Kit is a collection of programs written in Stata that provides some useful"
_n
"data management commands. Although Data.Kit is written in Stata's programming"
_n
"language, you do not have to know anything about the programming language to"
_n
"use Data.Kit. We begin by loading the kit:" _n(2) ;
#delimit cr
mac def path
capture run nullfile.tut
if _rc {
mac def path "\stata\"
capture run %path`nullfile.tut
if _rc {
mac def path "/usr/stata/"
capture run %path`nullfile.tut
if _rc {
#delimit ;
di in red
"I cannot find the other tutorial files. I have looked in the current" _n
"directory and in \stata (DOS) or /usr/stata (Unix). Is Stata installed" _n
"correctly?" _n(2)
"In any case, I cannot run the tutorial." ;
#delimit cr
exit
}
}
}
macro define F5 "do %path`contents.tut;"
macro define F6 "do %path`datakit.tut;"
drop _all
label drop _all
#delimit ;
di in wh ". run %path`Data.Kit" ;
program drop _all ;
run %path`Data.Kit ;
di ; set more 0 ; more ; set more 1 ;
di _n(2) in wh
"Overview" _n
"--------" _n ;
di in gr
"Data.Kit provides six new commands:" _n(2)
_col(8) in wh "collapse" in gr _col(20)
"make a data set of means, sums, or medians" _n
_col(8) in wh "order" in gr _col(20)
"change the order of the variables in a data set" _n
_col(8) in wh "recast" in gr _col(20) "change the type of a variable" _n
_col(8) in wh "mdytoe" in gr _col(20)
"change month/day/year to elapsed date" _n
_col(8) in wh "etomdy" in gr _col(20)
"change elapsed date into month/day/year" _n
_col(8) in wh "etodow" in gr _col(20)
"calculate day-of-week corresponding to elapsed date" _n ;
di in gr
"We will not demonstrate them all here. At the conclusion of this tutorial,"
_n
"you might type '"
in wh "help Data.Kit" in gr "' to learn more about the commands." ;
di _n in gr
"We were forced to omit the "
in wh "order" in gr " command from the demonstration version of" _n
"Data.Kit. During the process of making its calculations, "
in wh "order" in gr " writes a" _n
"temporary data set, something the demonstration version of Stata cannot do."
_n
"If you attempt to use "
in wh "order" in gr ", you will get the message '" in ye "not available in" _n
"demonstration version" in gr "'." _n(2)
"If you are using the Unix version of Stata in demonstration mode, do not use"
_n
"the " in wh "order" in gr
" command. Your version of Data.Kit contains the real " in wh "order"
in gr " program," _n
"but it will not work in demonstration mode." ;
set more 0 ; more ; set more 1 ;
di _n(9) in wh _dup(79) "-" _n in gr
"Data.Kit's " in wh "collapse" in gr
" command converts the data in memory into a data set of" _n
"means or medians or sums. To illustrate this, we'll use our Census data:" _n
in wh _dup(79) "-" _n(2)
". use %path`census.dta" ;
noisily use %path`census.dta ;
di _n in wh ". gen mrgrate = marriage/pop18p" ;
gen mrgrate = marriage/pop18p ;
di _n in wh ". gen dvcrate = divorce/pop18p" ;
gen dvcrate = divorce/pop18p ;
label var mrgrate "Marriage rate" ;
label var dvcrate "Divorce rate" ;
di _n in wh ". keep state region mrgrate dvcrate" ;
keep state region mrgrate dvcrate ;
di _n in wh ". describe" ;
set more 0 ; more ; set more 1 ;
noisily describe ;
di _n in wh ". list in 1/8" ;
noisily list in 1/8 ;
set more 0 ; more ; set more 1 ;
di _n(3) in wh _dup(79) "-" _n in gr
"If all we wanted was to see the means by region, we would not need Data.Kit."
_n
"Stata's built-in "
in wh "summarize" in gr " command combined with Stata's "
in wh "by" in gr " prefix is all we"
_n
"need:" _n
in wh _dup(79) "-" _n(2)
". sort region" ;
sort region ;
format mrgrate dvcrate %8.5f ;
di _n in wh
". by region: summarize mrgrate dvcrate" ;
set more 0 ; more ; set more 1 ;
noisily by region: summarize mrgrate dvcrate ;
set more 0 ; more ; set more 1 ;
di _n(2) in wh _dup(79) "-" _n in gr
"We, however, want to make a data set of means:" _n
in wh _dup(79) "-" _n(2)
". collapse mrgrate dvcrate, by(region) means" ;
noisily collapse mrgrate dvcrate, by(region) nowarning means ;
di _n in wh ". list" ;
noisily list ;
di ; set more 0 ; more ; set more 1 ;
di _n(6) in wh _dup(79) "-" _n
"mdytoe" in gr ", " in wh "etomdy" in gr ", and " in wh "etodow" in gr
" are three date manipulation commands. " in wh "mdytoe" in gr " con-"
_n
"verts a date in month/day/year format into an elapsed date, defined as the" _n
"number of days from "
in ye "January 1, 1960" in gr ". " in wh "etomdy"
in gr " converts an elapsed date back to" _n
"month/day/year format and "
in wh "etodow" in gr " derives the day-of-week (Sunday, Monday, Tues-" _n
"day, etc.) from an elapsed date. The syntax of the commands is:" _n ;
di _col(8) in wh "mdytoe " in gr "monthvar dayvar yearvar"
in wh ", generate(" in gr "elapvar" in wh ")" _n(2)
_col(8) "etomdy " in gr "elapvar" in wh ", generate("
in gr "monthvar dayvar yearvar" in wh ")" _n(2)
_col(8) "etodow " in gr "elapvar" in wh ", generate("
in gr "dowvar" in wh ")" _n(2) in gr
"This tutorial has silently loaded a data set containing some dates:" _n
in wh _dup(79) "-" _n(4)
". list" ;
drop _all ;
input str10 event int(month day year) ;
"Stata 1.0" 1 3 1985 ;
"Stata 1.1" 2 1 1985 ;
"Stata 1.2" 7 29 1985 ;
"Stata 1.3" 9 23 1985 ;
"Stata 1.4" 5 29 1986 ;
"Stata 1.5" 4 17 1987 ;
"Stata 2.0" 8 31 1988;
end ;
set more 0 ; more ; set more 1 ;
noisily list ;
di _n(3) in wh _dup(79) "-" _n in gr
"We can use " in wh "mdytoe" in gr" to create an elapsed-date variable:" _n
in wh _dup(79) "-" _n(3)
". mdytoe month day year, gen(elapsed)";
mdytoe month day year, gen(elapsed) ;
di _n in wh ". list" ;
set more 0 ; more ; set more 1 ;
noisily list ;
di _n(2) in wh _dup(79) "-" _n in gr
"We see that there are "
in ye "9,134" in gr " days between January 1, 1960 and the release of" _n
"the first version of Stata. There are "
in ye "10,470" in gr " days until the release of Stata"
_n
"2.0. Elapsed dates are more convenient than month/day/year format because you"
_n
"can subtract them. There are, for instance, "
in ye "10,470" in gr " - " in ye "9,134" in gr " = " in ye "1,336"
in gr " days be-" _n
"tween the release of Stata 1.0 and Stata 2.0." _n
in wh _dup(79) "-" _n ;
set more 0 ; more ; set more 1 ;
di _n(2) in wh _dup(79) "-" _n in gr
"Once we have the elapsed dates, there is no reason to keep the month/day/year"
_n
"dates. We can always get them back using " in wh "etomdy" in gr ":" _n
in wh _dup(79) "-" _n(4)
". drop month day year" _n(2)
". etomdy elapsed, generate(mo da yr)" ;
drop month day year ;
etomdy elapsed, generate(mo da yr) ;
di _n in wh ". list" ;
set more 0 ; more ; set more 1 ;
noisily list ;
di _n(2) in wh _dup(79) "-" _n in gr
"The " in wh "etodow" in gr
" command will derive the day-of-week from an elapsed date:" _n
in wh _dup(79) "-" _n(2)
". etodow elapsed, generate(wkday)" ;
etodow elapsed, generate(wkday) ;
di _n in wh ". list" ;
set more 0 ; more ; set more 1 ;
noisily list ;
di _n(6) ;
drop _all ;
label drop _all ;
macro define F6 "do %path`qckit.tut;" ;
set more 0 ; more ; set more 1 ;
di _n(4) in white
"Demonstration ends" _n
"------------------" _n ;
di in green
"That concludes our short demonstration, but there's much more. We now return"
_n
"control to you. Some suggestions:" _n ;
di in green
"If you ..." _col(34) "Then we will show you ..." _n
" Press " in white "F5" in green _col(38) "a table of tutorial contents" _n
" Press " in white "F6" in green _col(38) "the next tutorial, "
in white "qckit.tut" _n ;
run %path`tobuy.tut ;