I agree with all -- this is great! However, I think there is an error right at the end -- where you demonstrate how to enter an x value in the Basic Fitting window and click Evaluate, it provides an f(x) value. Then you select plot results. You say that it plots in the linear fit model. However, in the middle section of the Basic Fitting window, the selected Fit was "5th degree polynomial" altho display-selected was linear. "Evaluate" is relative to the mid-window Fit selected; so is plot.
I don't know of a simple way to do this. They will put the norm of the residuals on the plot. You can use this to calculate R^2 without too much difficulty.
That's a good question. I don't know of a simple way to do this. All I can think of is to take the log of the x-data, then plot and do the fit, and then take the exponential to get back to the normal axes.
i need to plot a graph using curvefitting. the signal is sound and is in db vs frequency. the origin graph was so pack.. how to fit it into smooth nice curve?
Thanks for the video. I am looking for a program that can fit 3D curves. Can this program do that? If not, can you recommend one? I would greatly appreciate it!
The curve fitting toolbox, purchased separately, will do a nonlinear fit of data to a surface. I would provide a link, but YouTube won't let me. Just google this
@jakeblanchard Thanks for the suggestion but when I google that all I get is this youtube page. ha!
Thanks for the suggestion though. If I find the answer I'll let you know so you can spread the word if anyone else comes to you with a similar question.
And what if I'd like to fit my points to custom function, like BirchMurnaghan equation of state? I'm currently searching for friendly way to do this fit...
My guess is that you need to do nonlinear regression. Take a look at the ezyfit package, which is free. It allows for custom functions in the fit. Just do a Google search for ezyfit.
thanks for the demo. I wish there were more like this. The basic curve fitting selection in the toolbox is nice but I need to use a squre root x function and I don't have the curve fitting utility.
You are doing a linear fit with an atypical function. LINEST in MS Excel does this fairly well. In Matlab, you might just want to treat it as a nonlinear fit. Take a look at this site:
This is awesome, I have spent like what? 2 months searching for a direct and easy way to do curve fitting method, and here it is, PERFECTLY EXPLAINED!!! Thanks Bro. Please any tips about contolling the way minor tick marks appear??? Thanks
thanx
AhmedTouma 2 months ago
I agree with all -- this is great! However, I think there is an error right at the end -- where you demonstrate how to enter an x value in the Basic Fitting window and click Evaluate, it provides an f(x) value. Then you select plot results. You say that it plots in the linear fit model. However, in the middle section of the Basic Fitting window, the selected Fit was "5th degree polynomial" altho display-selected was linear. "Evaluate" is relative to the mid-window Fit selected; so is plot.
ProfRHD 2 months ago
@ProfRHD
Nice catch. You are correct.
jakeblanchard 2 months ago
Is there a possibility to put R² on graph?
meteoroman100 4 months ago
@meteoroman100
I don't know of a simple way to do this. They will put the norm of the residuals on the plot. You can use this to calculate R^2 without too much difficulty.
jakeblanchard 4 months ago
thans for this video tutorial. It really helped me with my project. If you are interested i have some tutorials as well so go to my channel. Thanks
slash20able 8 months ago
Morgan Freeman!
retrogamingcables 1 year ago
@maezawi
That's a good question. I don't know of a simple way to do this. All I can think of is to take the log of the x-data, then plot and do the fit, and then take the exponential to get back to the normal axes.
jakeblanchard 1 year ago
Thanks so much for posting
anhthangyeu 1 year ago
Thank you ! This tip is very useful :)
qplaymygame 1 year ago
can i email the detail of my data to your email?
adyhao 1 year ago
@adyhao
Sure.
jakeblanchard 1 year ago
how to fit a sound signal? i mean if in db vs frequency the graph is too complicated. can i fit it into curve? thanks in advance!
adyhao 1 year ago
@adyhao
It's difficult to say without seeing the data. Is this a Bode plot?
jakeblanchard 1 year ago
i need to plot a graph using curvefitting. the signal is sound and is in db vs frequency. the origin graph was so pack.. how to fit it into smooth nice curve?
adyhao 1 year ago
it's visualized introduction
bmwmsn123 1 year ago
Thanks for the video. I am looking for a program that can fit 3D curves. Can this program do that? If not, can you recommend one? I would greatly appreciate it!
JarOfBuckeyes 1 year ago
@JarOfBuckeyes
The curve fitting toolbox, purchased separately, will do a nonlinear fit of data to a surface. I would provide a link, but YouTube won't let me. Just google this
Curve Fitting Toolbox 2.0 matlab
jakeblanchard 1 year ago
@jakeblanchard Thanks for the suggestion but when I google that all I get is this youtube page. ha!
Thanks for the suggestion though. If I find the answer I'll let you know so you can spread the word if anyone else comes to you with a similar question.
JarOfBuckeyes 1 year ago
@JarOfBuckeyes
OK. Try this.
mathworks "dot" com "slash" products "slash" curvefitting
You might need to add the three w's at the beginning
jakeblanchard 1 year ago
And what if I'd like to fit my points to custom function, like BirchMurnaghan equation of state? I'm currently searching for friendly way to do this fit...
zryw 2 years ago
My guess is that you need to do nonlinear regression. Take a look at the ezyfit package, which is free. It allows for custom functions in the fit. Just do a Google search for ezyfit.
jakeblanchard 2 years ago
nice,thanks heaps,man!
chriszeng8 2 years ago
i often find it easier to fit a 10th order polynomial by hand; matlab can be slow in these calculations
bobbybodickerson 2 years ago
thanks for the demo. I wish there were more like this. The basic curve fitting selection in the toolbox is nice but I need to use a squre root x function and I don't have the curve fitting utility.
Any suggestions?
JVLocascio 3 years ago
You are doing a linear fit with an atypical function. LINEST in MS Excel does this fairly well. In Matlab, you might just want to treat it as a nonlinear fit. Take a look at this site:
jakeblanchard 3 years ago
These youtube idiots won't let me post a comment with a URL in it. Let's try it this way:
blanchard . ep . wisc . edu
You'll have to take out the spaces. Then go to "Introduction to Matlab" and then "Curve Fits"
jakeblanchard 3 years ago
This is awesome, I have spent like what? 2 months searching for a direct and easy way to do curve fitting method, and here it is, PERFECTLY EXPLAINED!!! Thanks Bro. Please any tips about contolling the way minor tick marks appear??? Thanks
dabro4u 3 years ago 5
gut..ish bin gluchlish:)
syntheticsurfaces 3 years ago
Thanks
sdfg113s 3 years ago
You should give more online classes about Matlab. How can I contact you?
brenobfg 3 years ago
it is my first sentence that you must be more useful for people. much more!
You are good!. Thanks.
We will have been waiting for your Matlab's video.
muraday 4 years ago 2
good
hyuntaik 4 years ago