2D Fit Lines

You can fit an equation to the points in the 2D plots by selecting one of the predefined functions described below:

Linear. Select Linear to fit a linear function of the form:

Y = a + bX

to the points in the 2D plot. Confidence bands can be specified on the 2D Scatterplots - Advanced tab.

Polynomial. Select Polynomial to fit to the data a polynomial function of the form:

y = b0+ b1x + b2x2 + b3x3 +...+ bnxn

where n is the order of the polynomial (1<n<6). The order of the polynomial function can be changed in the Fitting dialog.

Logarithmic. Select Logarithmic to fit to the data, a logarithmic function of the form:

y = q*[logn(x)]+b

where the logarithm base (n) is selected by the user (by default, base 10 is selected). The base of the log function can be changed globally in the Documents/Graphs: Settings options pane in the Options dialog, accessible from the Tools menu. The log function can be changed for an individual graph on the Options 2 tab during graph creation.

Exponential function. Select Exponential to fit to the data an exponential function of the form:

y = b*exp(q*x)

where b and q are constants.

Least squares. Select Least Squares to fit a curve to the X-Y coordinate data according to the distance-weighted least squares smoothing procedure (the influence of individual points decreases with the horizontal distance from the respective points on the curve).

Negative Exponential. Select Negative Exponential to fit a curve to the X-Y coordinate data according to the negative exponentially weighted smoothing procedure (the influence of individual points decreases exponentially with the horizontal distance from the respective points on the curve).

Spline. Select Spline to fit a curve to the X-Y coordinate data using the bicubic spline smoothing procedure. For more information, see Spline Fitting.

Lowess smoothing. Select Lowess to use the Lowess method of smoothing data (pairs of X-Y data) in which a local regression model is fit to each point and the points close to it. The method is also sometimes referred to as robust locally weighted regression. The smoothed data usually provide a clearer picture of the overall shape of the relationship between the X and Y variables. For more information, see also Cleveland (1979, 1985). Note that the Lowess function in STATISTICA Visual Basic will also produce a table of residuals.