Ignore anomalies in data sets

Hello everyone

I want to import a data set and graph it, but after graphing I notice some anomalies in my data set causing an a number of values to significantly differ from the rest.

I'm wondering if there is a way I can ignore the faulty values to improve my graph's accuracy. Going through them one by one is impractical since there is more than 6000 cells to go though :( ...

Thank you very much

[Alex Mcaulley]

See rmoutliers

Answer 1

[Ted Shultz]

Do you have a rule that you can apply to define when a value should be ignored? (Above a certain value? distance from previous/next value?

If so, then the code could look like this:

%allData = readData
 
badDataIndex = allData > 10^4; % find locations of bad data
 
allData = allData(~badDataIndex); % drop any data that is bad

this could be done in one line as well:

allData = allData(allData<10^4); % drop any data that is above threshold