toolbar” near the middle or bottom of the screen. It may be necessary to “drag” the “plotting toolbar” up to
free some screen space below it after you complete the fitting process.
1. Linear fit using the “Tools” menu. The purpose of this section is not only to illustrate how to use the
“Tools menu” to obtain a least-squares fit of a data set but also to illustrate how to format the resulting
figure to conform to good scientific and engineering practices.
a. Open the figure saved above as ORIG1_A. (Click “File/Open/A/ORIG1_A”)’
b. With the plot in the forefront, use the “Format/Axes/X axis” sequence on the menu bar to
set the X axis range from 0 to 2.75.
c. Click on “Tools/Linear fit” from the menu bar. The drop-down “Fit” menu in the upper left
corner has two options, ”Operation” and ”Settings”.
d. Click on “Settings” first and set “Points” to 9 (there are nine sets of points in the data set),
“Confidence level” to 95, and “Range margin” to 0, clear “Residuals data” if it is set, and
click on the box “Span X Axis”.
e. Return to “Operation” and click on “Fit”; a fitted line should appear on your figure and
numerical values of least-squares statistics (slope, intercept, etc.) should appear below
the plotting toolbar. Click on the small x in the upper right corner of the “Linear fit” dialog
box to close it. If the numerical data for slope, intercept, etc. do not appear below the plot:
i. Click
on
“View/Results log”. If results still do not appear:
ii. Place the cursor on the bottom of the frame of the colorful toolbar at the bottom of the
screen and drag the toolbar up to about midscreen. If the numbers still do not
appear:
iii. Click
on
“View/Results log” again; fitting results should appear.
f. Click on the X in the upper right corner of the “Fit” dialog box to close it.
g. Save the figure on your disk as ORIG1_B.
h. After saving the figure, zoom in on it and note that the fitted curve passes through the
points; this is not good form. We will modify the figure later to pass the fitted curve “under”
the points.
7
7
i. Bring the plot to the forefront again.
2. Handling fitting parameters. The following procedure illustrates how to put the fitting parameters
(i.e. the numerical data below your figure) onto your figure and into a script window to be saved or
printed.
a. To put the fitting parameters onto your figure:
i. Put the cursor anywhere on the fitted parameters and click the RIGHT mouse button.
ii. Click “Select all”.
iii. Put the cursor anywhere on the fitting parameters again and click the RIGHT mouse
button; then click “Copy” in the dialog box (with the LEFT mouse button).
iv. Click “Edit/Paste” on the menu bar; the fitting parameters will be pasted onto your
figure. You can reduce the size by double clicking on the text and removing spaces
etc.
b. To put the fitting parameters into a script box:
i. Copy the parameters as described immediately above (a. i – iii).
ii. Click on “Window/Script window/Edit” from the drop-down menu.
iii. Click on “Edit/Paste” in the script window; parameters should be pasted into the
script window.
iv. Use the “File” function in the script window to save and/or print the fitting
parameters.
c. Save the figure as ORIG1_B and print the figure and script window.
3. Replotting the figure. It was noted above that it is bad form to have a fitted line overlay
experimental data. The following procedure will illustrate how the data points can be removed from the
graph and replotted over the fitted line.
a. With the figure in the forefront, click on “Data/Data A(X),B(Y)” from the menu bar.
b. With the “Data A(X),B(Y)” highlighted, press the delete key on the keyboard; data points
will disappear but the line will remain.
c. Double click on the small square with a number in it near the top left corner of the graph
frame.
d. Highlight (click on) “Data1 B” in the “Data available box”, click on the arrow (
⇒) toward the
“Layer contents” box and click on “OK”.
e. Zoom in on the resulting figure to observe that the fitted line does not pass through the
points.
f. Save the figure as ORIG1_B.
C. Fitting process – Custom model
Although Origin has many built-in fitting algorithims, you will need to use custom designed algorithms
from time to time. The linear data set is used here to illustrate how custom fitting programs can be used.
Close the active window, open the first window saved (e.g. ORIG_A) and be sure the plot is in the
forefront.
1. Creating a user-designed fitting algorithm. As noted earlier, the absorbance vs. concentration data
conform to a linear relationship of the form
A = a + bC (1b)
When custom fitting algorithms are used, Origin automatically assigns default parameter symbols, P1 and
P2, for the fitting parameters, unless one specifies otherwise. In this case, the resulting expression will
have the form
Y = P1 + P2*C (1c)
8
8
2. Selection of correct mode. The Origin software includes an “ Basic mode” and an “ Advanced
Dostları ilə paylaş: |