When I use quartile filter to remove outliers, how can I analitically express the thersehold factor? By default is 1.5, i need to know what rapresent thereshold factor of 4

% Remove outliers

[PV52021_nG_clPR,outlierIndices] = rmoutliers(PV52021_nG,"quartiles",...

"ThresholdFactor",4);

[Dyuman Joshi]

You have to use a two element vector - [40 60], instead of 4

rmoutlier threshold

[Gaetano Mannino]

Thanks for the answer, however when I use the task"clean outlier data" (in the livescript) i can chose the thereshold factor, i need to know analitically how limits are calculated (From Matlab documentation for quartile: Returns true for elements more than 1.5 interquartile ranges above the upper quartile (75 percent) or below the lower quartile (25 percent)..but when I increase the thereshold?

[Jonas]

you can read in the doc

Outliers are defined as elements more than 1.5 interquartile ranges above the upper quartile (75 percent) or below the lower quartile (25 percent). This method is useful when the data in A is not normally distributed.

there you can replace the 1.5 factor by 4 to understand what it means

[Gaetano Mannino]

thanks exactly, i would like to knoow how "upper and lower limits" are calculated to understand what happens changing the value of the thereshold factor (it can be 4 like the example or another value)

[Jonas]

someData=randn(10000,1)+randi(10);
quartiles=quantile(someData,[0.25 0.75]);
iqr=diff(quartiles);

factor=4;
upperBorder=quartiles(2)+factor*iqr

upperBorder = 16.0464

lowerBorder=quartiles(1)-factor*iqr

lowerBorder = 3.9419