Back to PLS Helperror bars on Excel - part 2
Posted on 06/26/08 12:32:36
Number of posts: 34
Nancy, thanks for the clarification (about the text box) and your reply regarding error bars on Excel. What I'm trying to do is to represent the behavior plots from PLS on Excel Spreadsheet. I transferred the correlation values for a particular LV to the spreadsheet and created a bar graph and now I would like to transfer the error bars from PLS to the Spreadsheet graph as well - is it possible for me to obtain the upper and lower values of the error bars on PLS (via Matlab or some other means)? Thanks again for your help!
Miyoung
Replies:
Untitled Post
I'm Online
Posted on 06/26/08 12:43:27
Number of posts: 229
quote:
Nancy, thanks for the clarification (about the text box) and your reply regarding error bars on Excel. What I'm trying to do is to represent the behavior plots from PLS on Excel Spreadsheet. I transferred the correlation values for a particular LV to the spreadsheet and created a bar graph and now I would like to transfer the error bars from PLS to the Spreadsheet graph as well - is it possible for me to obtain the upper and lower values of the error bars on PLS (via Matlab or some other means)? Thanks again for your help! Miyoung
All of the information you need is in the structure array labelled 'boot_result'Nancy, thanks for the clarification (about the text box) and your reply regarding error bars on Excel. What I'm trying to do is to represent the behavior plots from PLS on Excel Spreadsheet. I transferred the correlation values for a particular LV to the spreadsheet and created a bar graph and now I would like to transfer the error bars from PLS to the Spreadsheet graph as well - is it possible for me to obtain the upper and lower values of the error bars on PLS (via Matlab or some other means)? Thanks again for your help! Miyoung
You will want to save as ascii, or copy/paste into your worksheet the following:
boot_result.orig_corr
boot_result.ulcorr
boot_result.llcorr
each column reflects the results for an LV, each row is a behavioural measure (nested within conditions if >1 condition)
the ulcorr/llcorr is the value of the upper and lower confidence limit
you may also want to examine
boot_result.ulcorr_adj
boot_result.llcorr_adj
which give you estimated adjusted confidence intervals in the case of a large amount of skew in the bootstrapped distribution of correlation values.
Untitled Post
I'm Online
Posted on 06/26/08 12:44:57
Number of posts: 229
However, I do not think that Excel will let you plot errorbars where the upper and lower values differ across the datapoints ;-)
Login to reply to this topic.
