| Author |
 Topic  |
|
|
finesniper
Hungary
4 Posts |
 Posted - 08/11/2014 : 10:17:23 AM
|
Origin Pro 9. 64 bit Win 7
Hi everyone! I have a problem with two dimensional confidence interval fitting. I have two datasets in two columns. What I want to do is to check whether they belong to the same population, or form two distinct populations. For this, I created a two dimensional graph with the points, and I did the PCA, but I still want to look at the overlap of the two confidence interval. Where can I do this? I looked at the tutorial, but I could not find it. Thanks
Olh Viktor Jnos |
|
|
Echo_Chu
China
Posts |
Posted - 08/12/2014 : 05:15:22 AM
|
Hi,
What about using Mann-Whitney test, which is to test whether two population are teh same. You can access it by choosing menu Statistics: Nonparametric Tests: Mann-Whitney Test.
Echo
OriginLab Technical Service
|
 |
|
|
finesniper
Hungary
4 Posts |
Posted - 08/12/2014 : 2:22:41 PM
|
Thanks, that's a good way to solve my problem, but I still want to visualize the results.
Olh Viktor Jnos |
|
|
|
Echo_Chu
China
Posts |
Posted - 08/14/2014 : 10:14:54 PM
|
Usually in statistics if we want to see whether two population are the same from the graph, we will use the stacked histogram. Put the histogram of the two data together with same bin to observe whether they have same shape and location, dispersion and central tendency. In Origin you can plot such graph with the Statistics on Columns tool, accessible from Statistics: Statistics on Column
 |
|
|
| |
Topic |
|
Two dimensional confidence interval