How to select a random sample without replacement with imposed condition?

2 次查看（过去 30 天）

显示更早的评论

Christian 2015-10-24

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/250837-how-to-select-a-random-sample-without-replacement-with-imposed-condition

评论： Christian 2015-10-24

采纳的回答： Walter Roberson

在 MATLAB Online 中打开

Hi,

I'd like to draw a random sample from my population. My population is a vector x of size (N x 1). My sample should satisfy

sum(x(sample)) = 0.7*sum(x);

So I do not care how many observations I draw from the sample, but I just want to make sure that the resulting sample covers 70 percent of total x in terms of value, not in terms of number.

Thanks, Christian

2 个评论
显示无隐藏无

Walter Roberson 2015-10-24

Is that exactly 0.7 ? If so then it might not be possible to satisfy the condition.

Christian 2015-10-24

Good point. Let's say at least 0.7, but as close as possible to 0.7. So stop sampling as soon as the sum is equal or greater than 0.7.

请先登录，再进行评论。

请先登录，再回答此问题。

采纳的回答

Walter Roberson 2015-10-24

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/250837-how-to-select-a-random-sample-without-replacement-with-imposed-condition#answer_197211

在 MATLAB Online 中打开

rp = randperm(N);
loc = find( cumsum(x(rp)) >= 0.7 * sum(x), 1, 'first');
samps = x(rp(1:loc));

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Christian 2015-10-24

Thanks. Very neat.

请先登录，再进行评论。

类别

AI, Data Science, and Statistics Statistics and Machine Learning Toolbox Probability Distributions Continuous Distributions t Location-Scale Distribution

在 Help Center 和 File Exchange 中查找有关 t Location-Scale Distribution 的更多信息

Community Treasure Hunt

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

Start Hunting!

Translated by