fitting a smooth monotonic function to non noisy data
9 views (last 30 days)
Show older comments
Hi guys,
I'm new to this field of kernel estimation/smooth isotonic regression/smooth monotonic regression.
I need to fit a model for my strictly monotonic data,
I got lost with all the info online.
more specifficaly, I have data set points
, I know that the data is monoctonically increasing, no noise.
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/287943/image.png)
and I'm looking for a function f suct that
.
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/287944/image.png)
can you recommend a good algorithem to do so? or a paper to read, I want to understand how this process works.
In which conditions I will be able to recover the original function of the samples data?
Thanks!
Nushi
2 Comments
Ameer Hamza
on 27 Apr 2020
Do you have an equation of f(x), perhaps with few unknown parameters? You can use the data points to estimate the unknown parameters of f(x).
Adam Danz
on 27 Apr 2020
There are lots of monotonic functions that differ greately in their parameters. Consider a straight line vs an exponential curve. Both can be monotonically increasing. If you don't have the function f(x), perhaps you could plot the data points and share the plot so someone could suggest where to start.
Answers (0)
See Also
Categories
Find more on Linear and Nonlinear Regression in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!