. help fillin
Nick
[email protected]
[email protected]
My database includes the following variables:
- Country-pair
- Product lines
- Year
- Bilateral Trade flows;
- GDPs for exporter and importer countries;
- Preference tariffs;
The database treats as missing the observations where trade flows are
zero. I need to recover such zero values, and at the same time I have
to preserve the value of the other variables. Trying to explain with
an example, suppose having the following country pair : EU-ALBANIA
and EU-BARBADOS and suppose the total number of agricultural
product lines is 6 (actually they are hundreds, I'm just simplifying
to be clear). Now, in 2000 no lines is missing, by contrast in 2001
some product lines are missing as follows:
Country pair Lines Year Trade flows GDP_IMP GDP_EXP
Preference tariffs
EU-ALBANIA 1 2000 50 100 85 2
EU-ALBANIA 2 2000 60 100 85 3
EU-ALBANIA 3 2000 100 100 85 2
EU-ALBANIA 4 2000 90 100 85 1
EU-ALBANIA 5 2000 60 100 85 1
EU-ALBANIA 6 2000 40 100 85 7
EU-BARBADOS 1 2000 60 100 90 3
EU-BARBADOS 2 2000 100 100 90 2
EU-BARBADOS 3 2000 90 100 90 1
EU-BARBADOS 4 2000 60 100 90 1
EU-BARBADOS 5 2000 40 100 90 7
EU-BARBADOS 6 2000 40 100 90 6
EU-ALBANIA 1 2001 50 110 87 2
EU-ALBANIA 2 2001 90 110 87 1
EU-ALBANIA 4 2001 20 110 87 5
EU-BARBADOS 1 2001 50 110 92 2
EU-BARBADOS 6 2001 60 110 92 3
Instead I should have the following complete data for 2001
Country pair Lines Year Trade flows GDP_IMP GDP_EXP
Preferencetariffs
EU-ALBANIA 1 2001 50 110 87 2
EU-ALBANIA 2 2001 0 110 87 3
EU-ALBANIA 3 2001 0 110 87 2
EU-ALBANIA 4 2001 90 110 87 1
EU-ALBANIA 5 2001 0 110 87 1
EU-ALBANIA 6 2001 40 110 87 7
EU-BARBADOS 1 2001 60 110 92 3
EU-BARBADOS 2 2001 0 110 92 2
EU-BARBADOS 3 2001 90 110 92 1
EU-BARBADOS 4 2001 0 110 92 1
EU-BARBADOS 5 2001 0 110 92 7
EU-BARBADOS 6 2001 0 110 92 5.5
I have added extra rows so as to obtain 6 lines each year for each
pair of countries. Please notice that I need only the trade flows to
be zero, whilst the other variables have to maintain the same value
they have in that particular year. Moreover, I need to impute the
right line codes to the new created rows. This is a difficult task, as
there are so many different combinations of missing lines in different
years (the total number of lines is almost 500...)
I gave a look to the command FILL , but it does not seem to solve my
case. Do you know another command or way, which enables to get what I
need?
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
