am trying to find the first and second derivates of a set of data points I've imported

Hi,

I am trying to find the first and second derivates of a set of data points I've imported. I tried using the diff function but that just gave me an error. Also how would i create two seperate plots for each differentiation operation?

Thank you

[dpb]

Well, it would help at least some to know what the error was...and even more to know something about the data itself and the code you actually used...

[M]

the data im using is a file given to us for the assignment. it is a .csv file so i used M=readtable('filename.csv'). I then used y = diff(M). When i run it this is the error i get

Warning: Variable names were modified to make them valid MATLAB identifiers. The

original names are saved in the VariableDescriptions property.

Undefined function 'diff' for input arguments of type 'table'.

[KSSV]

What/How is the data you have in csv file?

[dpb]

Undefined function 'diff' for input arguments of type 'table'.

You insist on making it difficult by not providing sufficient information but M is the name of the overall table -- you've got to reference whatever particular variable it is out of the table to operate on...something like

dy=diff(M.X);

where X is the name of the variable you want to take difference of. Of, course, to actualy compute a derivative, you need the difference of both the displacement and time vectors if it were the velocity trying to compute--

v=diff(M.x)./diff(M.t);

[M]

i didnt know i could attach a file im not intending to make it difficult. i only have one row of data and i need the first and second derivatives of the data

Answer 1

[dpb]

That's essentially trivial file that

data=csvread('data_10.csv');
data(:,2)=[nan;diff(data(:,1))];          % first difference
data(:,3)=[nan(2,1); diff(data(:,1),2)];  % second difference

will leave you with an array of three columns for data, first and second differences, respectively. The difference vectors are each one element shorter than the previous of course, so I used NaN as filler so plotting routines will ignore them silently. Since they're columns, simply

plot(data)

will plot all three on one axes; you can select columns as desired and/or use two axes or whatever for aesthetics as desired.