I have a dataset of the following sort:

Code:
* Example generated by -dataex-. To install: ssc install dataex
clear
input str6 date long bns float bns_qrt_growth long(cbna fcib fgb)
"Mar-14" 176535310     . 11294717 35831053 22359818
"Jun-14" 190726980  8.04 10644840 35897166 23619625
"Sep-14" 192644381  1.01  9901093 37192064 23526022
"Dec-14" 194210805   .81 10754557 38620165 24288135
"Mar-15" 186943115 -3.74  9774468 40428726 25382380
"Jun-15" 198691679  6.28 13190912 43019852 26311049
"Sep-15" 214435176  7.92 13174967 44976924 26726495
"Dec-15" 212618211  -.85 13804100 50453410 26766775
"Mar-16" 220099378  3.52 12126162 54608312 29789672
"Jun-16" 244633857 11.15 11136283 51588411 32222677
"Sep-16" 237884451 -2.76 13277994 52869157 33116516
"Dec-16" 239354182   .62 13204974 51947427 33396243
"Mar-17" 243451013  1.71 12286569 54768652 32397767
"Jun-17" 258912892  6.35 14068066 58157548 35323039
"Sep-17" 262534320   1.4 15126217 58218143 39123263
"Dec-17" 259433254 -1.18 15284789 58249010 35861926
end
I calculated the quarterly growth rates of bns and put them under a new variable called bns_qrt. I want to do this for the rest of the variables [cbna, fcib, fgb] using a for loop as I have a lot more variables than these in my actual dataset. That's the first problem. To solve this problem, I was trying the following codes which resulted in an error - "f ambiguous abbreviation":

Code:
local banks cbna fcib fgb

local i = 1

foreach f of local banks {

        gen var`i' `f'_qrt_growth = (f - f[_n-1])*100/f[_n-1]

        local ++i
}


The second problem is, after calculating the quarterly growth rates, I want to divide these growth rates by 3 and then extrapolate monthly values for bns. For example, for June 2014, the bns growth rate is 8.04 which divided by 3 is 2.68. Now, for April 2014, the bns will be calculated as 176535310*(1+.0268).

If anybody could give some idea about how to go about it, that would be highly appreciated. I hope I am not asking for too much. Thanks.