extract

#! /bin/bash
#
# All files are downloaded directly from CDC website and are named
# accordingly to the year of reference. Column positions relevant
# to the data studied were checked manually for each year. This script
# extracts chese columns and saves them in smaller files that can then
# be treated with R. Some years have specific problems and have to be
# treated separately.

echo "$(cut -c38,39 1968)" >> "age_1968"
echo "$(cut -c47,48 1968)" >> "child_1968"

for i in {1969..1988}
do
	echo "$(cut -c41,42 $i)" >> "age_$i"
	echo "$(cut -c81 $i)" >> "child_$i"
done

for i in {1989..2002}
do
	echo "$(cut -c70,71 $i)" >> "age_$i"
	echo "$(cut -c201 $i)" >> "child_$i"
done

# Only recoded mother's age data available for 2003, otherwise it's unknown.
# Convert it back using R.

echo "$(cut -c89,90 2003)" >> "age_2003"
echo "$(cut -c423 2003)" >> "child_2003"

for i in {2004..2013}
do
	echo "$(cut -c89,90 $i)" >> "age_$i"
	echo "$(cut -c423 $i)" >> "child_$i"
done

for i in {2014..2015}
do
  echo "$(cut -c75,76 $i)" >> "age_$i"
  echo "$(cut -c454 $i)" >> "child_$i"
done


exit 0