Match Firms by SIC code (siccd) and Market Cap (marketcap)

[Bas Bosman]

Dear all,

We are trying to match a set of 1519 firms (ncusip_ipo) which IPO'd in 1975-1984, to a set of 5435 firms (ncusip_listed) which indicate all firms which were listed on the AMEX and NYSE in 1975-84, by matching according to 3 digit SIC codes (siccd) and market capitalization (marketcap). The matching firm selection procedure should be done as follows:

1. Within each 3-digit SIC code (siccd), the set of 5435 listed firms are ranked by (marketcap). For the 1519 IPO firms in a given 3-digit SIC code industry (siccd), the listed firm with the closest (marketcap) was chosen as the matching firm.
2. Furthermore, the matching firm can be used only once until 3 years have passed.

Can someone please help on both steps 1 and 2? Thanks in advance!

Regards,

Bas

[Phil Bromiley]

Welcome to Statalist. You didn't get a quick answer. You'll increase the chances of getting a useful answer by following the FAQ on asking questions – provide Stata code in code delimiters, readable Stata output, and sample data using dataex. This also ensures that you have actually tried to solve their own problem forest asking for help.

You obviously should begin by sorting by SIC code and market cap. Then the matching firm should be the one above or below a given firm, although you may need to look further than just above and below sinse you might have an IPO firm next to an IPO firm by marketcap. After you've done this, you will probably find that a given firm has been matched to more than one of the IPO firms - you may need to go back and fix these manually. the other way to avoid duplicate matches would be to loop over your IPO firms, but I think this is more trouble than it's worth.