| T O P I C R E V I E W |
| diallos |
Posted - 06/07/2024 : 3:43:55 PM
This is for a Physics research data analysis. So basically:
I have some data points I want to curve fit in LabOrigin.
Fitting all the points gives not so great values. so we tried fitting all the points before 200k which gave us better output values.
To show that the fitting isn't great for temperatures above 200k, my research advisor wants me to include all the data points in the graph, but mask all data points above 200k. That is, graph all data points, fit for points below 200k, mask those above 200k, but still have points showing in graph (even masked ones)
I found few solutions online but none of them that I tried give me the results we are trying to get. This may be because none of the points are outliers (outside the fitting line). I attached a picture below for what the graph should look after the fitting.
Does anyone know of any solutions I could try that would make the graph look like the picture in LabOrigin? I have not being having any luck finding anything online/research papers.
Thank you in advance. |
| 4 L A T E S T R E P L I E S (Newest First) |
| NicholasSupport |
Posted - 07/02/2024 : 5:14:58 PM
Hello Salima,
I apologize for the delayed reply. I missed your response. I hope it is not too late for input!
It looks like you are looking at the Fitted Curves Plot in FitLinearN worksheet? Since masked data gets excluded from analyses, the Fitted Curves Plot in FitLinearN worksheet will never include any masked data points.
To achieve your desired effect, you must first plot the data, mask the data, then fit the data.
Nicholas G.
nicholas@originlab.com
OriginLab Technical Support Team |
| diallos |
Posted - 06/07/2024 : 5:27:24 PM
@NicholasSupport. I have already tried the solution you mentioned. The problem is when I click recalculate, it automatically removes the masked data points from the graph and instead looks like this. Which is not what we are going for. 
|
| NicholasSupport |
Posted - 06/07/2024 : 5:18:04 PM
Hello Salima,
I think the best approach is to click and activate your graph window within your project.
Select the Mask Points tool from the toolbar. The flyout menu for this button allows users to mask/demask points on the active plot or all plots in a graph. Make sure you select the best one for your project.
You can control the shape of your selection by tapping the spacebar. Pressing tab allows you to cycle through the active plot.
Create a range around the data you would like to exclude from analysis. You may have to press the Recalculation button to view your updated results.
Here is a webpage that you can use for reference on this feature.
https://www.originlab.com/doc/Quick-Help/Exclude-Outliers (Text)
https://www.originlab.com/videos/details.aspx?id=261 (Video)
I am also attaching a project file so you can play around.
https://my.originlab.com/ftp/forum_and_kbase/Images/MaskGraphExample.zip
Best,
Nicholas G.
nicholas@originlab.com
OriginLab Technical Support Team |
| ChaoC |
Posted - 06/07/2024 : 4:48:42 PM
Hello,
You can directly mask the >200K points in your worksheet or graph.
The analysis tools, such as fitting, will exclude the masked data.
https://www.originlab.com/doc/Origin-Help/Mask-Data
https://www.originlab.com/videos/details.aspx?id=261
Best,
Chao |
|
|
