olap - Calculating PERCENTILE in DAX -

i have googled , keep ending formulas slow. suspect if split formula in steps (creating calculated columns), might see performance gain.

i have table having numeric columns along end slicers. intention have 10th, 25th, 50th, 75th , 90th percentile on numeric columns selected slicer.

this have 10th percentile on column "total pd".

totalpaid10thpercentile:=minx(                               filter(                                      values(claimoutcomes[total pd]),                                      calculate(                                                countrows(claimoutcomes),                                                claimoutcomes[total pd] <= earlier(claimoutcomes[total pd])                                                )> countrows(claimoutcomes)*0.1                                     ),                                claimoutcomes[total pd]                              ) 

it takes several minutes , still no data shows up. have around 300k records in table.

i figured out way break calculation down in series of steps, fetched pretty fast solution.

for calculating 10th percentile on amount paid in table data, followed below out-of-the-book formula :

calculate ordinal rank 10th percentile element

10ptordinalrank:=0.10*(countx('data', [amount paid]) - 1) + 1 

it might come out decimal(fraction) number 112.45

compute decimal part

10ptdecpart:=[10ptordinalrank] - trunc([10ptordinalrank]) 

compute ordinal rank of element below(floor)

10ptfloorelementrank:=floor([10ptordinalrank],1) 

compute ordinal rank of element above(ceiling)

10ptceilingelementrank:=ceiling([10ptordinalrank], 1) 

compute element corresponding floor

10ptfloorelement:=maxx(topn([10ptfloorelementrank], 'data',[amount paid],1), [amount paid]) 

compute element corresponding ceiling

10ptceilingelement:=maxx(topn([10ptceilingelementrank], 'data',[amount paid],1), [amount paid]) 

compute percentile value

10thpercvalue:=[10ptfloorelement] + [10ptdecpart]*([10ptceilingelement]-[10ptfloorelement]) 

i have found performance remarkably faster other solutions found on net. hope helps in future.