back-transform rank data

10 次查看(过去 30 天)
locas
locas 2018-4-6
i used tiedrank function to rank my data in order to convert them into standard normal distribution.
rank=tiedrank(Data);
p= rank/(length(rank) +1 );
newdata=norminv(p,0,1);
now i need to backward transform newdata to its original dist. Any suggestions ?
Thanks.

回答(1 个)

Jeff Miller
Jeff Miller 2018-4-7
I am not really sure what you are trying to accomplish since you still have the original Data, but maybe this will give what you are after:
backtrans = mean(Data) + std(Data)*newdata;
Alternatively, I think you could get the same thing directly with:
newdataback=norminv(p,mean(Data),std(Data));

类别

Help CenterFile Exchange 中查找有关 Spline Postprocessing 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