Negative values in lognormal distribution (lognfit)

4 次查看(过去 30 天)
I have a 1 x 4000 vector which also contains some negative values. The shape seems to resemble a log normal distribution but the lognfit will not work because of the negative values. Please suggest a workaround for this. What distribution should I use, if not lognormal? Or if I want to continue using lognormal how to do it?
Thanks,
  1 个评论
Jan
Jan 2017-7-12
The question is not clear, because the readers cannot guess how the negative values should be treated. If you get them from a valid measurement, and do not want to ignore them, there is no way to fit them to a distribution, which does not allow negative values. There is not even a "workaround" for this, because you ask for something, which is not mathematically meaningful. Then even posting multiple threads for the same question does not help.

请先登录,再进行评论。

回答(0 个)

Community Treasure Hunt

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

Start Hunting!

Translated by