Dear stata community

I have to do a fixed effect panel regression with the difference in ask yield of a bond as y_{i,t} on the difference in bid ask spread x_{i,t}, this a measure for liquidity.

∆y ̃i,t = pi +β∆Liquidityi,t +εi,t

The difference in ask yield is the difference of a certain bond against 2 other bonds issued by the same issuer and the bid ask spread is that the same bond against the same 2 other bonds. the 3 bonds form a triplet together. I have data on 170 differences in ask yield on differences in bid ask spread over a time period between 2015 until 2018. The method i use is that of: Is there a Green Bond Premium? The yield differential between green and conventional bonds by Olivier David Zerbib (2018) https://papers.ssrn.com/sol3/papers....act_id=2889690

The problem is that i dont understand how i can do a regression off all the y_{i,t} on all the x_{i,t} while still matching the correct y with the correct x. I think i will need to build a matrix to do this but i do not know how that works in stata. How do i have to prepare my data so that i can do a fixed effect panel regression and what do i have to type in the command section to get the answer? At the moment i am working with an excel file where i have daily values of the differences in ask yield and daily values of bid ask spreads over time for each triplet of bonds. Each triplet has a different starting point because the issue date is different.