Hello.
Can we regard the correlation plot between two parameters using winBUGS
as the true simulated correlation plots between the two parameters ?
In other words, when we calculate the value of correlation coefficient
between two parameters using samples generated by winBUGS, can we regard
the value as a estimate of the correlation coefficient ?
I wonder if autocorrelations between samples are reflected in the plots.
Please tell me .
Hidetoki Ishii
[log in to unmask]
--------------------------------------
Department of Educational Psychology
Graduate School of Education
The University of Tokyo
--------------------------------------
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|