a data set in Matlab?

I have a data set that I need to graph and then find the area under the curve.

2/12/2007 9:28:25 PM

i just looked under the help, searcher "integrate data"

Quote :

"Can I integrate a set of sampled data? Not directly. You have to represent the data as a function by interpolation or some other scheme for fitting data. The smoothness of this function is critical. A piecewise polynomial fit like a spline can look smooth to the eye, but rough to a solver; the solver takes small steps where the derivatives of the fit have jumps. Either use a smooth function to represent the data or use one of the lower order solvers (ode23, ode23s, ode23t, ode23tb) that is less sensitive to this."

but i would think there is another way

perhaps you can calculate your area by summing the y-values multiplied by the step width in your x-values?

EDIT:

the function "trapz" looks to be the one, check it out in the help

[Edited on February 12, 2007 at 9:37 PM. Reason : trapz]

2/12/2007 9:35:04 PM

well shit, i keep getting an error when i try to run matlab through putty

and I need this before I can finish my lab report that is due tomorrow

[Edited on February 12, 2007 at 9:59 PM. Reason : ]

2/12/2007 9:49:24 PM

might need a trip to broughton tonight

if its just plotting the data and getting the area, send me the data and your code to plot/calculate it, and i'll run it and paste it in word and send it back. rstruzik@gmail.com

2/12/2007 10:02:12 PM

Matlab typically runs through GUI, so it won't work directly on Putty. If you run XWin with Putty, Matlab will usually be fine.

2/12/2007 10:05:47 PM

I think you might be right about that trip to broughton . . . although i'm not sure what time those labs close now

I would take you up on your offer, but I don't even have that code in front of me.


^I am running X-win32 and Putty. It says "Fatal Java Exception Detected"

[Edited on February 12, 2007 at 10:08 PM. Reason : ]

2/12/2007 10:06:37 PM

^ that shit always lagged like hell for me, if you're close enough it's easier to just go to campus

edit:

i think broughton closes at 11

dh hill is open 24hrs mon-thurs

or laundry labs should be open all night i think

[Edited on February 12, 2007 at 10:12 PM. Reason : times]

2/12/2007 10:09:52 PM

newton's method.

2/12/2007 10:16:23 PM

Daniels had the 24hr engineering computer lab last i checked, though that was a year ago

[Edited on February 12, 2007 at 10:18 PM. Reason : on the 1st or 2nd floor]

2/12/2007 10:18:02 PM

You could always just use excel to plot the data and use Simpson's rule for the area unless you have a lot of data points.

2/12/2007 10:21:20 PM

^I have 12 data points . . . what is simpson's rule again?

2/12/2007 10:34:22 PM

%Define the function to integrate (using anonymous functions)
f = @(x) x^2;

%Set the interval to integrate
a = -1;
b = 1;
%set the number of panels to compute and their length
n = 100;
h = (b-a)/n;

%split up the interval into subintervals
x = [a:h:b];
%note that matlab matrices are indexed starting at 1
sum = f(x(1));
for i=2:2:n
    sum = sum + 4*f(x(i));
end;
for i=3:2:n-1
    sum = sum + 2*f(x(i));
end;
%prints out the result
f_integrated = (h/3)*(sum + f(x(n+1)))

2/12/2007 10:37:32 PM

yeah, I remembered Simpsons rule shortly after I posted that

I just did it in excel real quick . . . not hte most accurate, but it'll do for now


/thread



[Edited on February 12, 2007 at 10:43 PM. Reason : thx everyone]

2/12/2007 10:42:44 PM

um...

if you have a discrete set of points, you don't really integrate that. the discrete version of an integral is just a summation. like deltax*sumofdata. if you have an explicit that you want matlab to numerically integrate, than you define the function either as an m file or as an anonymous function and then use the quad command:

help quad

which uses i think gaussian quadratures to evaluate. simpsons rule uses parabolic quads. anyways..have fun.

2/13/2007 4:32:06 PM