function - Algorithm for determining acceptable variance -


i have application comparing numbers 2 different reports. going alert user when variance between 2 numbers on threshold. going use 10% threshhold, realized when count ex. 10,000 variance of 10% high (meaning if there 999 excessive), when count 10 variance of 10% low (meaning 2-3 / 10 acceptable variance).

just can't figure out how it, besides coding out

if counta <= 10 acceptablerate = 20% if counta > 10 acceptablerate = 15% 

does know how explain trying here mathematically, , how implemented? sure simple question better @ math am.

besides formulas mentioned in previous answers, consider using power of n (where n reference number 1 report or other) tolerance. here python code , results, illustrating several different powers of n:

j=10    in range(6):     print '{:8} {:8.1f} {:8.1f} {:8.1f} {:8.1f}'.format(j, j**0.33, j**.35, j**.37, j**.39)     j *= 10        10      2.1      2.2      2.3      2.5      100      4.6      5.0      5.5      6.0     1000      9.8     11.2     12.9     14.8    10000     20.9     25.1     30.2     36.3   100000     44.7     56.2     70.8     89.1  1000000     95.5    125.9    166.0    218.8 

Comments

Popular posts from this blog

jquery - How can I dynamically add a browser tab? -

node.js - Getting the socket id,user id pair of a logged in user(s) -

keyboard - C++ GetAsyncKeyState alternative -