diff --git a/UK-IPS-time-series-extract.do b/UK-IPS-time-series-extract.do
index 1047c11388a35dc53635105d163deceee1ab97b0..510d7866ddba8fad6d00e67391461198045328f5 100644
--- a/UK-IPS-time-series-extract.do
+++ b/UK-IPS-time-series-extract.do
@@ -45,8 +45,10 @@ set more off
* set these to what works for you
-local place = "/Users/ben/Documents/Work"
local ipsd = "`place'/Data/Social Science Datatsets/UK International Passenger Survey"
-local logd = "`ipsd'/log_files"
local outd = "`ipsd'/processed"
+local place = "~/Documents/Work"
+local ipsd = "`place'/Data/Social Science Datatsets/UK International Passenger Survey"
+local logd = "`ipsd'/log_files"
+local outd = "`ipsd'/processed"
************************
* NB - the script assumes a certain folder structure for the source IPS data like so:
@@ -83,7 +85,7 @@ local qcont_keepvars = "year qtr month serial flow age purp fweight"
* set logging
capture log close
log using "`logd'/UK-IPS-time-series-extract-$S_DATE.smcl", replace
-
clear all
+clear all
* increase default number of variables allowed
set maxvar 10000, perm
@@ -95,7 +97,7 @@ di "* It will keep the following variables:"
di "* `keepvars' "
di "* This could take some time. "
di "* I suggest you check it is running and then get a cup of tea...."
-di "*******************************************************"
+di "*******************************************************"
if `do_extracts' {
di "do_extracts = `do_extracts', all years (`years') to be extracted and refreshed"
@@ -103,8 +105,10 @@ if `do_extracts' {
foreach y of local do_years {
di "* * * * * * * * * "
di "* -> Processing `y'"
-
******************************
- * qcount
di "* qcount file"
+
+ ******************************
+ * qcount
+ di "* qcount file"
local pref = "qcont"
local suff "`y'"
@@ -140,33 +144,62 @@ if `do_extracts' {
* in some years there are 4 quarterly files
if "`y'" == "01" | "`y'" == "02" | "`y'" == "03" | "`y'" == "04" | "`y'" == "05" | "`y'" == "06" | "`y'" == "09" | "`y'" == "12" | "`y'" == "14" {
local qs "1 2 3 4"
- foreach n of local qs {
use "`ipsd'/20`y'/stata/`pref'`n'`suff'.dta", clear
* make everything lower case
+ foreach n of local qs {
+ use "`ipsd'/20`y'/stata/`pref'`n'`suff'.dta", clear
+ * make everything lower case
rename *, lower
- * save all vars - process later
save "`outd'/tmp/qcont`n'`y'-temp.dta", replace
******************************
+ * save all vars - process later
+ save "`outd'/tmp/qcont`n'`y'-temp.dta", replace
+ ******************************
}
di "*** APPEND ALL FILES for year = `y'"
*
- clear
qui: append using ///
"`outd'/tmp/qcont1`y'-temp.dta" ///
- "`outd'/tmp/qcont2`y'-temp.dta" ///
"`outd'/tmp/qcont3`y'-temp.dta" ///
- "`outd'/tmp/qcont4`y'-temp.dta"
+ clear
+ qui: append using ///
+ "`outd'/tmp/qcont1`y'-temp.dta" ///
+ "`outd'/tmp/qcont2`y'-temp.dta" ///
+ "`outd'/tmp/qcont3`y'-temp.dta" ///
+ "`outd'/tmp/qcont4`y'-temp.dta"
* DELETE TEMPORARY FILES
- foreach n of local qs {
+ foreach n of local qs {
erase "`outd'/tmp/qcont`n'`y'-temp.dta"
- }
}
else {
+ }
+
+ }
+ else {
use "`ipsd'/20`y'/stata/`pref'`suff'.dta", clear
* make everything lower case
rename *, lower
}
- ** sex
gen ba_gender = 0 if sex == 1
replace ba_gender = 1 if sex == 2
- replace ba_gender = . if sex == 9
lab var ba_gender "Gender of HRP"
lab def ba_gender 0 "Male" 1 "Female"
lab val ba_gender ba_gender
** age
- recode age (0=0) (2=1) (4 = 2) (5 = 3) (6 = 4) ///
(7 = 5) (8 = 6) (else=.), gen(ba_age) // codes don't know & youth group parties to missing
lab var ba_age "Age group (individual)"
- label define ba_age ///
0 "0-15" ///
1 "16-24" ///
2 "25-34" ///
3 "35-44" ///
4 "45-54" ///
5 "55-64" ///
6 "65+"
lab val ba_age ba_age
* check
+ ** sex
+ gen ba_gender = 0 if sex == 1
+ replace ba_gender = 1 if sex == 2
+ replace ba_gender = . if sex == 9
+ lab var ba_gender "Gender of HRP"
+ lab def ba_gender 0 "Male" 1 "Female"
+ lab val ba_gender ba_gender
+
+ ** age
+ recode age (0=0) (2=1) (4 = 2) (5 = 3) (6 = 4) ///
+ (7 = 5) (8 = 6) (else=.), gen(ba_age) // codes don't know & youth group parties to missing
+ lab var ba_age "Age group (individual)"
+ label define ba_age ///
+ 0 "0-15" ///
+ 1 "16-24" ///
+ 2 "25-34" ///
+ 3 "35-44" ///
+ 4 "45-54" ///
+ 5 "55-64" ///
+ 6 "65+"
+
+ lab val ba_age ba_age
+ * check
*tab age ba_age, mi
/*
@@ -228,7 +261,7 @@ if `do_extracts' {
keep `kvars' ba_*
qui: compress
-
+
gen survey_year = "20`y'"
tab survey_year
@@ -236,9 +269,15 @@ if `do_extracts' {
* keep all (makes life easier below as some vars are in some years and not others)
/* create a birth cohort variable
- Birth year Age in 2000 Age in 2010
1900-1909 90-100
1910-1919 80-90 90-100
1920-1929 70-80 80-90
1930-1939 60-70 70-80
1940-1949 50-60 60-70
1950-1959 40-50 50-60
+ Birth year Age in 2000 Age in 2010
+ 1900-1909 90-100
+ 1910-1919 80-90 90-100
+ 1920-1929 70-80 80-90
+ 1930-1939 60-70 70-80
+ 1940-1949 50-60 60-70
+ 1950-1959 40-50 50-60
1960-1969 30-40 40-50
- */
+ */
* remember these are based on age groups
* set the minimum age they could be
recode ba_age (0=0) (1=16) (2=25) (3=35) (4=45) (5=55) (6=65), gen(ba_min_age)
@@ -256,7 +295,8 @@ if `do_extracts' {
drop ba_birthyear ba_min_age
qui: compress
save "`outd'/IPS-20`y'-extract-BA.dta", replace
- local count = "`count' + 1"
}
+ local count = "`count' + 1"
+ }
}
****************************
@@ -295,6 +335,6 @@ else
{
di "*-> do_extracts = `do_extracts', years not extracted so individual files not refreshed"
}
-
di "*-> Job ended at $S_DATE"
+di "*-> Job ended at $S_DATE"
-log close
+log close