| T O P I C R E V I E W |
| tashara |
Posted - 06/23/2025 : 09:34:58 AM
Origin Ver. and Service Release (Select Help-->About Origin):
Operating System:
I would like to ask, previously I was still a beginner to do this and I would be very grateful if you could teach me. I have data on microfibril and pore diameters, here I want to see the distribution and average of microfibril and pore diameters. From the results of the histogram plot, my data looks skewed, possibly using log-normal. After I did log-normal and measured the mean with the calculation formula (mean = exp (μ + σ² / 2)) but the results were too large for microfibrils> 1µm, while my data did not have microfibrils that big, all 0,.. µm. The steps I took are listed below:
1. data from imagej (length and feret diameter are entered into the originpro table)
2. each data is made a histogram
3. Histogram is autobinned
4. I made the bin results into a column
5. Then I analyzed -> fitting -> nonlinear curvit -> open log -> category (peak function) and function (lognormal)
6. the results are listed in the picture
are the steps I took wrong? or is there an explanation? does my data allow to use gaussian or is it really lognormal? do I need to transform the data? then the average microfibrils and pores that I want come from the initial lognormal or the result of the data transformation and then fit with gaussian?
tasya |
| 3 L A T E S T R E P L I E S (Newest First) |
| YimingChen |
Posted - 06/24/2025 : 09:14:35 AM
They should give you same result. In your case:
mean = exp (μ + σ² / 2) = exp(-2.06215 + 0.42478^2/2)=0.139
Yiming |
| tashara |
Posted - 06/24/2025 : 12:57:01 AM
Thank you sir for your answer,, before that I would like to ask again, sorry here I am still new to learning it and a little confused. For the average (here I want to know the average pore), for the average pore I use the average and std from descriptive statistics, or do I have to go through the mu and sigma calculations obtained from the lognormal?because the mean results are different
tasya |
| YimingChen |
Posted - 06/23/2025 : 5:27:25 PM
Please use the Distribution Fit tool (menu Statistics: Descriptive Statistics: Distribution Fit...) to fit your raw data. Fitting the distribution to the histogram is not recommended. For more information, please refer to this FAQ:
https://www.originlab.com/doc/Quick-Help/CurveFit-on-Histogram
Yiming |
|
|
