Outliers search in script

[Eni]

Hi, I'm new. I'm trying to learn qlik to use it at work... Just now I'm using qlik cloud but I'll use qlik enterprise in the future so I need general solutions and just script-related since I will be dealing with massive databases.
For now...
I'm trying to detect outliers but I think the problems is more with my comprehension of expansions. 



test:
LOAD
codfis,
imposta1,
tassa2,
grandezza3,
valore150
FROM [lib://DataFiles/tabellatest.xlsx]
(ooxml, embedded labels, table is Foglio1);

let campo = 'codfis';
let tabella = 'test';
let id = '$(tabella)'&'$(campo)';

sub proc($(id),$(tabella),$(campo))

//the mistake is in the following statement, if I don't use expansions but codfis, imposta1 etc it works just fine. I tried using let statements before but it returns an error (as it were nothing in their place) when I try to load them

diff'$(camp)':

load (fractile($(camp), 0.99)- fractile($(camp), 0.25)+1) as dif'$(camp)'

resident '$(tabella)';

outlier:
load

count($(campo)) as numero resident '$(tabella)'where $(campo) > 1000*dif'$(camp)';
end sub;
call out ($(camp));

let camp = imposta1;
call out ($(camp));
end sub;


I tried again doing it in the big for each cycle... 
Now the first part works (Idk why, I just copied the one that didn't) but then the error message board tells me that it can't find the field 'a' (... I have no field a, I checked the grafical variables too) and shows me "where $(campo) > 1000*$(dif);" as "where $(campo) > 1000* "

test:
LOAD
codfis,
imposta1,
tassa2,
grandezza3,
valore150
FROM [lib://DataFiles/tabellatest.xlsx]
(ooxml, embedded labels, table is Foglio1);

let campo = 'codfis';
let tabella = 'test';
let id = '$(tabella)'&'$(campo)';

sub proc($(id),$(tabella),$(campo))

for each campo in 'codfis','imposta1','tassa2','grandezza3','valore150';
let tabella = 'test';
let id = '$(tabella)''_''$(campo)';

diff: load (fractile($(campo), 0.99)- fractile($(campo), 0.25)+1) as dif resident '$(tabella)';
outliers: load '$(id)' AS id, '$(tabella)' as tabella, '$(campo)' as campo, count($(campo)) as numero resident '$(tabella)' where $(campo) > 1000*$(dif);
next;
end sub;