using fold to conditionally filter a set of numbers for averaging

[wlbsr]

is there a way filter the fold loop to leave out the greatest "m" values of the 1 to n set of "ret" values. ie add an m input where if m=1, leave out the one largest value when averaging the sum of the "ret" variables (divide the sum by n-m). Thanks in advance.

declare lower;


input n = 30;


def ret=absValue(close/close[1]-1);


plot mad = fold i = 0 to n
                    with s = 0
                   do s + GetValue(ret, i, n) / n;

 

[halcyonguy]

is there a way filter the fold loop to leave out the greatest "m" values of the 1 to n set of "ret" values. ie add an m input where if m=1, leave out the one largest value when averaging the sum of the "ret" variables (divide the sum by n-m). Thanks in advance.

declare lower;


input n = 30;


def ret=absValue(close/close[1]-1);


plot mad = fold i = 0 to n
                    with s = 0
                   do s + GetValue(ret, i, n) / n;

i've seen a post that removes the biggest and smallest values... somewhere but i can't find it.
maybe it's related to standard deviation or the mean or numbers, or...

--------------------

EDIT

i think that just study i referenced above just listed the highest... so it wouldn't help in your case.

i made a study to find the mth highest number in a range of numbers.
next i will add code to find an average.


TSLA

 

[halcyonguy]

is there a way filter the fold loop to leave out the greatest "m" values of the 1 to n set of "ret" values. ie add an m input where if m=1, leave out the one largest value when averaging the sum of the "ret" variables (divide the sum by n-m). Thanks in advance.

declare lower;


input n = 30;


def ret=absValue(close/close[1]-1);


plot mad = fold i = 0 to n
                    with s = 0
                   do s + GetValue(ret, i, n) / n;

here is a study that does most of what you were asking for
https://usethinkscript.com/threads/ignore-extreme-values-when-calculating-an-average.14364/