XY
The xy plotting and analysis window is opened by choosing the XY menu
option from the NMRView control panel. Any data accessible with the command
interpreter in NMRView can be placed in a data array that
can be plotted in this XY plotting window. Nonlinear
regression routines are also available to fit the data
with a variety of equations commonly used in NMR
analysis. Any of the Meschach type vectors that contain real or complex
data may be plotted. By default, the xy routines plot data with the
x values taken from the real vector fit_x and the y values taken
from the real vector fit_y. These same vectors are used for the
non-linear regression analysis. Additionally, the standard deviations are
taken from the real vector fit_s.
When you choose the xy option from the analysis menu
a new window will open, as displayed below.
The menubar at the top has the following functions:
- Display
- Click this button to refresh the xy plot. Bug Note: In the current
version, the plot initialy appears as the small box in the upper left corner.
Clicking on Display will fix this.
- AutoScale
- Click this button to automatically rescale the plot limits and refresh
the plot.
- Plot
- Click this button to create a file containig a Postscript representation
of the plot.
- Fit
- Click this button to fit a model equation to the data using a non-linear
curve fitting algorithm. The algorithm, equation and various parameters
are specified in the FitPar panel. See the documentation for the FitPar
panel for more information on the methodology.
- FitPar
- Click this button to open a control panel with various parameters
controlling the curve fitting process.
- Close
- Click this button to close the XY panel.
- Equation
- Pull down menu used to select the desired equation to fit
to the data.
- NSim
- Use the slider to specify the number of simulated data sets to be
used in the Monte-Carlo method for determining confidence intervals.
- Clv
- Use the slider to specify the confidence level used
in the Monte-Carlo method for determining confidence intervals.
- Fitting Method
- Specify the algorithm used for fitting the data.
- Fit
- This command causes a non-linear curve fit to
be done to the data stored in the xy data array. The
equation fitted is that specified in the equation pull
down menu. The xy data and a line corresponding to the
fitted parameters are displayed on the xy graph. The
best fit parameters are displayed in the xy control
window. Confidence intervals are calculated using a Monte-Carlo method
similar to that
described in Numerical Recipes (Press et al.). The data is fitted once, and then simulated data is generated
by adding gaussian noise to data simulated with the best fit parameters.
This simulated data is then fit to the same equation, with the best fit
parameters as starting values. The process is repeated (typically 100 times).
The array of values for each parameter is sorted and the two values (min and
max) are chosen that bracket a specified percentage of the data. These values
are reported in the Min and Max fields for each parameter.
- Data
- The fitted data is output in the console window. The Best, Max, and Min
values for each parameter are reported.
- Best
- These entries are the best fit values to each of the parameters
in the equation (A,B,C ...).
- Min
- The lower value for the specified confidence interval.
- Max
- The upper value for the specified confidence interval.
- Close
- Close the fitPar panel.