| Author |
 Topic  |
|
|
dhpii
Denmark
3 Posts |
 Posted - 05/21/2011 : 09:38:16 AM
|
Origin Ver. and Service Release (Select Help-->About Origin): OriginPro 8.5.0 (acedemic)
Operating System: Windows seven
Hi All,
Please help me out with these two problems:
1)
How do you make a Logarithmic fit (regression) on a dataset in a plot with log-scale X-axis and linear Y-axis. The equation should be y = b*ln(x) + a. I know this can be done in Excel but I'm lost in Origin.
2)
I need to change my Y-axis from being left to right (w/o adding an extra Y-axis) - how can this be done?
I should mention that I'm new at Origin, so please bear with me!
Thanks!
David |
|
|
dhpii
Denmark
3 Posts |
Posted - 05/21/2011 : 09:41:25 AM
|
| *Service release: SR1 |
 |
|
|
Hideo Fujii
USA
1582 Posts |
Posted - 05/23/2011 : 10:02:43 AM
|
Hi David,
For 1), as your fitting function is linear on log scale, you can perform the linear fit ("Analysis: Fitting: Linear Fit" menu) on the log-scaled plot. In the opened dialog, you can turn ON the check box of "Apparent Fit".
For 2), double-click your Y-axis, choose "Title&Format" tab, turn OFF the "Show Axis&Ticks" check box, select "Right" axis selection (icon), and turn ON the "Show Axis&Ticks" check box.
--Hideo Fujii
OriginLab |
Edited by - Hideo Fujii on 05/23/2011 10:08:32 AM |
|
|
|
dhpii
Denmark
3 Posts |
Posted - 05/29/2011 : 6:52:35 PM
|
Hi Hideo Fujii,
Thanks for your help! It is really appreciated.
Though by doing a linear fit on the log scale I don't get the same slope and interception as I do by doing the same thing in MS Excel - that's why I wondered whether the linear fit could be used as a logarithmic regression.
BR
David |
|
|
|
ccu
7 Posts |
Posted - 07/20/2011 : 03:32:21 AM
|
| What is the meaning of "Apparent Fit" and what's the different with/without it? I don't really understand the explanation in help file. |
|
|
|
Hideo Fujii
USA
1582 Posts |
Posted - 07/20/2011 : 3:53:34 PM
|
quote:
Originally posted by ccu
What is the meaning of "Apparent Fit" and what's the different with/without it? I don't really understand the explanation in help file.
Hi,
"Apparent Fit" means the fitting is performed "as it appears" on a non-linear scale. For example semi-log-on-X plot, the apparent fitting is based on the linear relationship: y = a + b*log(x) as shown in the screenshot below (both graphs are identical except linear X scale at right):
Does it make sense?
--Hideo Fujii
OriginLab |
|
|
|
ccu
7 Posts |
Posted - 07/23/2011 : 11:21:38 AM
|
| I see...this setting regard the log scale as linear scale to perform linear fitting... Thanks for your effort! |
|
|
| |
Topic |
|
Logarithmic regression and right Y-axis only