Finding confidence intervals on fitted parameters
3 ビュー (過去 30 日間)
古いコメントを表示
Hi there.
I have some sets of 1-dimensional experimental data, that I fit to a rather complicated function, with up to 9 different free parameters, in addition to some fixed parameters, that vary between data sets. Some of the parameters are unfortunately somewhat correlated.
I fit the data using a package called spec1d: http://www.ill.eu/instruments-support/computing-for-science/cs-software/all-software/matlab-ill/spec1d/ I have been using it for a long time, and would rather not switch to another set of tools, as I have to hand in a thesis soon, and don't think I have time to rewrite all my code. However, I am lacking something from the program. I would like to calculate the 68% and 95% confidence interval of my parameters as they illustrate in fig. 15.6.4 in Numerical Recipes (<http://www.fing.edu.uy/if/cursos/fiscomp/extras/numrec/book/f15.pdf)>.
I never did an analysis like this before; usually it is a fair assumption that the parameters are uncorrelated, and the program automatically gives reasonable error bars.
So if anyone can help me out I would greatly appreciate it! I hope I provided enough information, if not then please ask away.
0 件のコメント
回答 (0 件)
参考
カテゴリ
Help Center および File Exchange で Fit Postprocessing についてさらに検索
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!