Problems with removing data

[Qi vd Kolk]

Hi all,

I'm using Stata 17 on MacBook.
For my Master Thesis I have a panel data set where I have accounting data for the years 2016-2020 for US firms.
However, after dropping missing values, some firms end up not having data for the years 2016-2020 but for instance only for the years 2019 and 2020.
Below is a screenshot of a part of my data where a firm that only has data for the years 2019 and 2020 is highlighted. How do I drop the firms that do not have data for all the years (2016-2020) that I need?

Thanks in advance!

[Andrew Musau]

15 posts in and you are still posting screenshots. Please review the FAQ Advice to see how to present data examples in the forum.

*Ensure no duplicates
isid cusip fyear

*Check no years outside 2016-2020
assert inrange(fyear, 2016, 2020)

*Wanted
bys cusip: keep if _N== 5

[Qi vd Kolk]

Hi Andrew,

Thank you very much!
I posted a screenshot because It didn't work when I tried to post a small sample of my dataset as described in the FAQ.

When I use your code it only keeps the firms with data for years 2016-2020.
However, the years are no longer in order from 2016 to 2020 but mixed up (e.g. the order 2017,2019,2020,2016) for each company.
I tried to use the command 'egen' to fix this problem, but it doesn't work.

Do you maybe know how I can get it back in the order from 2016 to 2020?

[Andrew Musau]

I posted a screenshot because It didn't work when I tried to post a small sample of my dataset as described in the FAQ.

You can restrict the data. See

help dataex

In your example:

dataex datadate-cusip in 115/140

will provide a data example of your screenshot in #1.

Do you maybe know how I can get it back in the order from 2016 to 2020?

sort cusip fyear

See

help sort

[Qi vd Kolk]

Thank you very much Andrew, both codes work!