I have an array of data [ 1.1 0.5 1.8 2.5 5.0 4.2 3.3 4.8 2.5] After the first number (1.1) need pick the next higher number which is 1.8 until until you get to the peak (5.0) then pick the number in descending order from the peak to the lowest

I have an array of data [ 1.1 0.5 1.8 2.5 5.0 4.2 3.3 4.8 2.5] After the first number (1.1) I need to pick the next higher number which is 1.8 until you get to the peak (5.0) then pick the number in descending order from the peak again by picking the next next lower number after 5.0 which is 4.2 follow by 3.3 then drop 4.8 because is increasing not decreasing until you get to the last number which is 2.5. how to code this

[Selby]

Here have a clunky answer!

dat = [ 1.1 0.5 1.8 2.5 5.0 4.2 3.3 4.8 2.5];
[max_value,max_position] = max(dat);
%claculate first part
   first_part = dat(1:max_position);
   first_part_differences = diff(first_part);
   increasing_logical = first_part_differences>=0;
   increasing_part = [first_part(1),intersect(first_part,first_part(2:end).*increasing_logical)]
%lastpart
   last_part = dat(max_position:end);
   last_part_differences = diff(last_part);
   decreasing_logical = last_part_differences<=0;
   decreasing_part = fliplr([intersect(last_part,last_part(2:end).*decreasing_logical)]);
answer = [increasing_part,decreasing_part];

[the cyclist]

If I interpret Olumide's question correctly, your algorithm will not do what they want on the vector

dat = [ 1.1 0.5 1.8 1.1 1.7 2.5 5.0 4.2 3.3 4.8 2.5];

[Selby]

my understanding is that he wanted the numbers increasing to the max then decreasing afterward such that the output would be [1.1 1.8 2.5 5.0 4.2 3.3 2.5] maybe it excludes the 1.1 but i think it was left fairly ambiguous

[the cyclist]

Agreed with the first part, but your algorithm gives

answer =

    1.1000    1.7000    1.8000    2.5000    5.0000    4.2000    3.3000    2.5000

[Selby]

Oh I see! Sorry for confusion. I didn't read your comment fully this should work better.

    dat = [ 1.1 0.5 1.8 2.5 5.0 4.2 3.3 4.8 2.5];
    [max_value,max_position] = max(dat);
    first_part = [dat(1)];
    current_highest = dat(1);
    for i=2:max_position
        if dat(i)> current_highest
            first_part = [first_part dat(i)];
            current_highest = dat(i);
        end
    end
    last_part = [];
    current_lowest = max_value;
    for i=max_position+1:length(dat)
        if dat(i)< current_lowest
            last_part = [last_part dat(i)];
            current_lowest = dat(i);
        end
    end
    answer  = [first_part last_part];

[Stephen23]

See also (almost) identical question (with answers):

https://www.mathworks.com/matlabcentral/answers/352074-i-have-an-array-of-data-1-1-0-5-1-8-2-5-0-6-2-0-2-1-4-0-1-3-1-4-5-0-4-2-2-2-3-2-3-4-3-5-3-3-4-8-1?s_tid=prof_contriblnk

Answer 1

[Jan]

x = [1.1 0.5 1.8 2.5 5.0 4.2 3.3 4.8 2.5];

% Find the maximum at first:
[v, idx] = max(x);

% Then run two loops to filter out the cumulative maximum/minimum:
result    = zeros(1, numel(x));  % Pre-allocate
result(1) = x(1);
c         = 1;
for k = 2:idx
  if x(k) > result(c)
    c         = c + 1;
    result(c) = x(k);
  end
end

for k = idx+1:numel(x)
  if x(k) < result(c)
    c         = c + 1;
    result(c) = x(k);
  end
end

result = result(1:c);   % Crop unused elements

Answer 2

[Andrei Bobrov]

data = [1.1 0.5 1.8 2.5 5.0 4.2 3.3 4.8 2.5];

[~,ix] = max(data);
ii = false(size(data));
r = data(1);
for jj = 1:numel(data)
    if jj <= ix && r <= data(jj)
        ii(jj) = true;
        r = data(jj);
    elseif jj >= ix && r >= data(jj)
        ii(jj) = true;
        r = data(jj);
    end
end
out = data(ii)