Hello everyone, i am new to the forum so i hope my post follows the recommendatios of the forum. I have a monthly panel data with 36 banks from January 2011- December 2019 and i want to estimate the effect of implementing negative interest rates on bank lending and to include time and bank fixed effects. I have a variable called RD_ratio (ratio of Retail deposits to total assets) and i want to divide my banks in two groups, high_RD_ratio and low_RD_ratio by using the median to divide them, in a similar way as stated in the paper of Demiralp, S., Eisenschmidt, J. and Vlassopoulos, T., 2019 where they "classify banks into three business models: high, medium, and low retail deposit intensity, based on the size of their “retail deposits/ assets” ratios, using the 33rd and 66th percentiles to split the sample".

My data look something like this:

dataex bankid month total_assets NIRP total_lending RD_ratio Loan_ratio
Code:
* Example generated by -dataex-. To install: ssc install dataex
clear
input byte bankid long(month total_assets) float(NIRP total_lending RD_ratio Loan_ratio)
8 201908 91347 1 74910  .1608044 .8200598
8 201909 89939 1 75539 .16214323 .8398915
8 201910 89680 1 75946  .1624554 .8468555
8 201911 90025 1 76347 .16088864 .8480644
8 201912 88467 1 76441 .16332644 .8640623
9 201101 71749 0 27850  .5381817 .3881587
9 201102 69826 0 28198  .5554951 .4038324
9 201103 70633 0 28612  .5559724 .4050798
9 201104 71745 0 28980 .55228937 .4039306
9 201105 71891 0 29429 .55294824 .4093558
end
I would like some with with the code if possible in order to produce a table that looks like this one from the paper. I assume i should use either xtqreg or qregpd but i am not sure what is the best choice and how the code should be.
Thank you in advance for any help!

Array