Home / Statistical Tools / Analysis Tools / Scatterplot / Options

Options¶

The Scatter Plot dialog lets you choose a regression model, configure extrapolation, and apply data transformations.

Regression¶

Controls the type of regression line fitted to the data. The regression equation, R², and other statistics are displayed alongside the chart.

None¶

Plots data points only with no trendline or regression statistics.

Linear¶

Fits a straight line: y = a + bx. This is the default and best for relationships where Y changes at a constant rate as X changes.

Logarithmic¶

Fits a logarithmic curve: y = a + b·ln(x). Requires positive X values. Best when the rate of change in Y decreases as X increases.

Polynomial¶

Fits a polynomial curve of order 2 through 6. Use the order spinner to select the polynomial degree. Higher orders can fit more complex curves but risk overfitting — a model that fits the sample data well but predicts poorly on new data.

Power¶

Fits a power curve: y = a·x^b. Requires positive X and Y values. Best for relationships where Y changes proportionally to a power of X.

Exponential¶

Fits an exponential curve: y = a·e^(bx). Best for relationships showing exponential growth or decay.

Extrapolation¶

Extrapolate Model¶

When checked, extends the regression line beyond the observed data range as a dotted line. The solid line shows the fitted model within the data range, and the dotted segments show predictions outside it.

Extrapolate Backward to X¶

The X value where backward extrapolation starts. The dotted line extends from this value to the minimum observed X value. Leave empty to skip backward extrapolation.

Extrapolate Forward to X¶

The X value where forward extrapolation ends. The dotted line extends from the maximum observed X value to this value. Leave empty to skip forward extrapolation.

Data Transformation¶

Controls how X values are transformed before regression fitting. Transformation applies only to X values and can improve regression fit when X values have very different scales.

Option	Description
None	Use raw X values as-is (default)
Code	Transforms X values to a -1 to +1 scale. Useful when X values span a large range.
Standardize	Transforms X values to z-scores (subtract mean, divide by standard deviation). Useful when comparing regression across datasets with different X scales.