Graph Plotting Pane
To obtain information about the Graph Pane’s controls, mouse-click them in the figure shown below.
Tip: Double-click the data entry field with the mouse to maximize its size and again to minimize it. This enables long coordinate series to be entered more easily.
Data are entered as coordinate pairs, in the field at the top-left of the graph plotting pane, in the following format:
x-value followed by a comma and then the y-value – eg. 3,2
Each data pair must be entered on a new line: if you enter data wrongly, a dialogue box will display requesting that the data be re-entered.
You can also use the keypad to enter data.
Data can be entered as decimals numbers or in standard-from:
For example: 430.6 or 4.306E-2
Data are sorted, as you enter them, in x-value ascending order.
Tip: Numerical values can be entered as decimals or in scientific form:
eg. 42.05 or 4.205e1 0.0056 or 5.6e-3
Positive and negative numbers in the range 10 -300 to 10 300 are supported.
Setting either the x or y axis type to ‘Date & Time’ or ‘Time’ causes the data for that axis to be interpreted as either calendar or 24-hour clock time, as the case may be.
Calendar Entries
Date and time values are entered or edited by accessing either the pop-up menu for the data entry field or by pressing the letter ‘D’ on the keyboard when the cursor is located in this field. A dialogue box is then displayed that allows you to enter a date and time value in the correct calendar format.
Calendar Entry Dialogue Box
NB. Date and time entries are rendered as decimal data values in accordance with the following scheme:
0 = 30th December 1899 at 00.00 hours
1 = 31st December 1899 at "
2 = 1st January 1899 "
-1 = 29th December 1899 "
38353 = 1st January 2005 "
Hours, minutes, seconds and milliseconds are determined by the decimal places in the entered value:
For example 12:00 on the 1st January 2005 is entered as 38353.5
Tip: When using calendar or time axes, it is best to set the auto-scale facility. This avoids your having to manually convert scale axes ranges from date and time values to a decimal values.
When an axis type is set to Time, data are entered or edited by accessing either the pop-up menu for the data entry field or by pressing the letter ‘T’ on the keyboard when the cursor is located in this field. A dialogue box is then displayed that allows you to enter a time value in the correct format:
NB. Time entries are rendered as decimal data values which correspond to the number of seconds since the beginning of the day:
12:00 hour = 43200 seconds
12:30 = 45000 "
12:30:30 = 45030 "
Entered data can be edited by typing directly into to the data entry field or by using the keypad.
Tip: Don’t forget that, for calendar or time axes, you should use the data entry dialogue box instead of typing directly into the data entry field.
To plot a graph from data contained in an external file, use the ‘Graph Options’ menu and select ‘Load Coordinates’. In the dialogue box that displays next, you can select the data file that you want to load.
To save an existing data set, repeat the above but choose the ‘Save Coordinates’ option instead.
NB. Data loaded from file must be formatted as comma-delimited x and y value pairs on separate lines, ie. with a carriage return character at the end of each line, and must be saved with a filename that has ‘DTA’ as its extension (eg MyDate.DTA).
Tip: You can also select these menu options by accessing the data entry field’s pop-up menu.
Coordinates can also be entered by clicking the left mouse button at the location within the chart at which you want to add a data point.
Data can be deleted by removing the entry from the data entry field itself or by placing the mouse cursor close to the point within the chart and left-clicking the mouse whilst the keyboard’s shift key is depressed.
Tip: If you have difficulty getting the mouse to delete data points, try increasing the value in the ‘Precision’ box located beneath the chart.
Error-bars can be individually assigned to y-values or set generically as a y-value percentage, the y-value standard error or in accordance with a use-defined function.
Error-bar options are displayed in the chart’s pop-up menu. Here you can choose how the error-bar is displayed.
Setting error-bars individually
Each coordinate can be assigned a positive and negative error by adding these values inside square brackets:
For example: x,y[positive error value][negative error value]
Tip: If you want to use the same value throughout the plotted data, set the error-bar values in the first coordinate pair and then select the ‘Use Previous’ option in the chart’s pop-up menu.
Setting error-bars to a percentage of the x-value.
Select the ‘Use Percentage’ option in the chart’s pop-up menu and then enter a percentage value in the dialogue box.
Setting error-bars to the standard error
Select the ‘Use Standard Error’ option in the chart’s pop-up menu.
Select the ‘Use Custom’ option in the chart’s pop-up menu and then enter a function in the dialogue box:
For example: sqrt(y) adds error-bars as the square-root of the y-value.
Text labels can be added to charts by either using the chart’s pop-up menu and selecting the ‘Add Label’ option or by right-clicking the mouse at the location within the chart at which you want the label added.
A maximum of ten labels are permitted.
Tip: Labels can be moved by selecting the ‘Drag Label’ option in the ’ pane. When this option is selected, labels can be dragged with the mouse by clicking on them and holding down the left mouse button as the mouse is moved.
Double-clicking a label successively changes the display format: text, underlined, boxed and shadow-boxed options are available. If you have difficulty selecting a label, increase the value of the precision shown in the control underneath the chart.
Labels can be deleted by left-clicking the mouse on them whilst, at the same time, holding down the shift key on the keyboard.
Tip: If you have difficulty selecting a label, increase the value of the precision shown in the control underneath the chart.
Plotted data can be fitted to number of regression functions by selecting an entry in the ‘Curve Fitting Functions’ control. Parameters relating to the selected function are displayed in the fields to the right of this control.
Tip: Double-mouse click any of the line-fit calculated parameters (see labels in the top-right of the diagram below) to turn on or off the extension of the regression curve through the y-axis (ie. y-axis intercept plotting).
Line-fit Settings and Parameters
Calculates the best-fit straight line to a sample of two-dimensional data points using linear regression. The line is defined by the equation:
y = kx + d
A minimum of two coordinate x/y pairs are required by this fitting function.
The line-fit function returns the parameters: k and d, which define the slope and the y-axis intercept of the line, and "Fit Quality" which defines the goodness-of-fit of the regression line to the data. "Fit Quality" is numerically equal to the square of the correlation coefficient and is equal to unity where the fitting quality is perfect.
NB. The quality of fit calculated by this fitting function is not adjusted for the degrees of freedom in the regression data (ie. the error-bars).
Calculates the best-fit logarithmic curve for a given set of data. The curve is determined by the equation:
y = k0 + k1 * ln(x)
The values of x and y are given by the data samples, the parameters k0 and k1 are estimated by the fitting function using a least squares approximation.
In addition to the parameters k0 and k1, the fitting function calculates the quality-of-fit of the regression curve. This parameter may vary between 0.0 and 1.0: a value of unity indicates the best possible fit.
NB. The quality of fit calculated by this fitting function is not adjusted for the degrees of freedom in the regression data (ie. the error-bars).
Calculates the best-fit gaussian curve (normal distribution) for a given set of data. The curve is determined by the equation:
- [(x-k1) 2 / k2 )]
Y = k0e
The values of x and y are given by the data samples, the parameters k0, k1 and k2 are estimated by the fitting function using a least squares approximation. A minimum of 3 values are required in order to apply this fitting function.
In addition to the parameters k0, k1 and k2, the fitting function returns a goodness-of-fit value. This parameter may vary between 0.0 and 1.0: a value of unity indicates the best possible fit.
NB. The quality of fit calculated by this fitting function is not adjusted for the degrees of freedom in the regression data (ie. the error-bars).
Calculates the best-fit hyperbola for a given set of data. The curve is determined by the equation:
Y = k0+( k1/x )
The values of x and y are given by the data samples, the parameters k0 and k1 are estimated by the fitting function using a least squares approximation. A minimum of 2 values are required in order to apply this fitting function.
In addition to the parameters k0 and k1, the fitting function returns a goodness-of-fit value. This parameter may vary between 0.0 and 1.0: a value of unity indicates the best possible fit.
NB. The quality of fit calculated by this fitting function is not adjusted for the degrees of freedom in the regression data (ie. the error-bars).
Calculates the best-fit parabola for a given set of data. The parabola is determined by the equation:
Y = k0+k1x+k2x2
The values of x and y are given by the data samples; the parameters k0, k1, and k2 are estimated by the fitting function using a least squares approximation.
A minimum number of 3 values are required in order to apply this fitting function.
In addition to the parameters k0, k1, and k2, the fitting function returns a goodness-of-fit value. This parameter may vary between 0.0 and 1.0: a value of unity indicates the best possible fit.
NB. The quality of fit calculated by this fitting function is not adjusted for the degrees of freedom in the regression data (ie. the error-bars).
Calculates the best-fit reciprocal curve for a given set of data. The curve is determined by the equation:
Y = 1 / ( k0+k1x )
The values of x and y are given by the data samples, the parameters k0 and k1 are estimated by the fitting function using a least squares approximation. A minimum of 2 values are required in order to apply this fitting function.
In addition to the parameters k0 and k1, the fitting function returns a goodness-of-fit value. This parameter may vary between 0.0 and 1.0: a value of unity indicates the best possible fit.
NB. The quality of fit calculated by this fitting function is not adjusted for the degrees of freedom in the regression data (ie. the error-bars).
Cubic spline "fitting" calculates the best-fit function ( y=f(x) ) that passes through all data points in the plot.
The calculated function has a smooth first derivative and a continuous second derivative.
Tip: On older computers, line-fitting may take a long time to be updated when the chart is formatted or when plotted data are changed. Therefore, this option should only be enabled when you have finished preparing the chart and are ready to either print or save it to file.