Optimization -Estimate optimal coeffiecients

4 次查看（过去 30 天）
Mon Mart
Mon Mart 2017-4-13
You are interested in modeling the shape of an individual action potential using the function,
f(x)= b(1)/(1+exp.^(-b(2)*(x-b(3))))- b(1)/(1+exp.^(-b(4)*(x-b(5))));
ActionPotentialRecords.mat, contain the voltage traces of several observed action potentials in the APs variable, as well as the time of each observation in times, whose values are [1260x1 double]. Estimate the optimal coefficients of your model based on the data available in the file and use the method of undersampling to determine the coefficient of variation (confidence measure) of each of your fit parameters (use 30 data subsets). An often used confidence threshold is 0.05 for the coefficient of variation.
if true
% code
end
% 1) Create an anonymous function, fx, of your model
fx = @(x,p) ???;
for k=1:30
% 2) Create an array of indices, ix, that represent 1/30th of your data
% and that change with each loop iteration, k
ix = ???;
% 3) create your objective function (sum of squared error)
SSEf = @(x) sum( ( ??? ).^2 );
% 4) perform the optimization
???
end
% 5) compute parameter means and standard deviations

请先登录，再回答此问题。

回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Nonlinear Optimization 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by