Find minimum or maximum value - two conditions

Hi there,

I have a problem where I am trying to find the index of minimum or maximum values based on two conditions. I have separated my data into 10 second blocks, with 1 data point per second. The two conditions are as follows:

1) the next minimum value must not be within 5 seconds of the previous minimum value

2) the next minimum value must be within the 10 second block following the previous minimum value's 10 second block.

Presently, I can ensure the next indexed minimum value meets one condition or the other, but I cannot ensure that it meets both.

Any ideas would be much appreciated!

Luke

[Image Analyst]

Have you tried a for loop? You forgot to attach your data and expected result, which would enable us to provide a solution to you sooner.

If you have any more questions, then attach your data and code to read it in with the paperclip icon after you read this:

TUTORIAL: How to ask a question (on Answers) and get a fast answer

[LukeJes]

Please see attached.

I've included my code below, up until where I am having the problem.

My expected result would be a minimum value from each 10 second block (odds:evens) that are each not within 5 seconds of their previous minimum value. Ideally it'd be a find function with two conditions, such as find(min(data(idxmcamin(j-1)+5:evens(j),2)) && min(data(odds(j-1):evens(j),2)). I hope this makes sense!

clear
spreadsheets = dir('*.xlsx');
spreadsheets = {spreadsheets.name};
for i = 1:length(spreadsheets)
    trial = spreadsheets(i);
    trial = char(trial);
    data = xlsread(trial);
    evens = 0:10:length(data);
    evens = evens(1,2:end);
    odds = 1:10:length(data);
    odds = odds(1,2:end);
    mcaoutliers = isoutlier(data(:,2));
    mapoutliers = isoutlier(data(:,3));
    if sum(mcaoutliers)>0
        data(:,2) = filloutliers(data(:,2),"linear");
    elseif sum(mapoutliers)>0
        data(:,3) = filloutliers(data(:,3),"linear");
    else
        %
    end
    % find max and min and time
    for j = 1:length(evens)
        [idxevens,~] = find(data(:,1)==evens(j));
        if j == 1
            % mca
            valmcamin(j) = min(data(2:idxevens,2));
            valmcamax(j) = max(data(2:idxevens,2));
            [idxmcamin(j),~] = find(data(:,2)==valmcamin(j),1,'first');
            [idxmcamax(j),~] = find(data(:,2)==valmcamax(j),1,'first');
            timemcamin(j) = data(idxmcamin,1);
            timemcamax(j) = data(idxmcamax,1);
            % map
            valmapmin(j) = min(data(2:idxevens,3));
            valmapmax(j) = max(data(2:idxevens,3));
            [idxmapmin(j),~] = find(data(:,3)==valmapmin(j),1,'first');
            [idxmapmax(j),~] = find(data(:,3)==valmapmax(j),1,'first');
            timemapmin(j) = data(idxmapmin,1);
            timemapmax(j) = data(idxmapmax,1);
        else
            % mca
            valmcamin(j) = min(data(idxmcamin(j-1)+5:evens(j),2)); % PROBLEM
            valmcamax(j) = max(data(idxmcamax(j-1)+5:evens(j),2)); % PROBLEM
            [idxmcamin(j),~] = find(data(:,2)==valmcamin(j),1,'first');
            [idxmcamax(j),~] = find(data(:,2)==valmcamax(j),1,'first');
            timemcamin(j) = data(idxmcamin(j),1);
            timemcamax(j) = data(idxmcamax(j),1);

[LukeJes]

Please let me know if you require any other information to help with answering this question :)

Answer 1

[Steven Lord]

I think the islocalmin and islocalmax functions will be of use to you. See the MinSeparation name-value argument.

[LukeJes]

Hi Steven,

Thanks for the suggestion. Unfortunately islocalmin with the MinSeparation argument is a bit off in places. I've attached my data and code above to give you more idea of what I'm working with :)

[LukeJes]

Please let me know if you require any other information to help with answering this question :)