How can I create a zero-truncated binomial distribution pf?
3 ビュー (過去 30 日間)
古いコメントを表示
Hi all,
The page http://www.mathworks.com/help/stats/examples/fitting-custom-univariate-distributions.html shows how to make a zero-truncated Poisson probability function for MLE fitting. Does anyone know how to do the same thing for a binomial distribution?
I have the following code:
function paramests = fitBinoMixture(hist)
N = 10;
pdf_binomixture = @(x, t, p1, p2)...
t*binopdf(x,N,p1) + (1-t)*binopdf(x,N,p2);
%Parameters
t_start = 0.5;
p1_start = 0.03;
p2_start = 0.5;
start = [t_start p1_start p2_start];
lb = [0 0 0];
ub = [1 1 1];
%Rum MLE
paramests = mle(hist, 'pdf',pdf_binomixture, 'start',start, 'lower',lb, 'upper',ub);
This works on simulated data, but not on my experimental data in which zero-occurrence events are not observed.
Thanks for any advice! Dan
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 Probability Distributions 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!