Xojo Developer Conference
25/27th April 2018 in Denver.
MBS Xojo Conference
6/7th September 2018 in Munich, Germany.

Platforms to show: All Mac Windows Linux Cross-Platform

CDTrendLayerMBS class

Super class: CDLayerMBS

Type Topic Plugin Version macOS Windows Linux Console & Web iOS
class ChartDirector MBS ChartDirector Plugin 8.2 Yes Yes Yes Yes No
Function: The TrendLayer class represents trend layers.
The trend layer performs linear regression analysis on the data points, and represents the result as a best fit straight line with optional confidence and prediction bands.

In linear regression analysis, the data points are assumed to be related by:

y = m * x + c + err
where m and c are constants, and err is a random variable.

Linear regression analysis estimates m, c and err based on available data using the least square method. Using estimated values of m and c, the line y = m * x + c are plotted as the best fit straight line based on available data.

However, as m and c are estimations based on available data, it may not be exactly equal to the "real" m and c. In ChartDirector, the uncertainties are represented visually as a confidence band around the regression line. For example, the 95% confidence band means there are 95% probability that the "real" line is in that band.

To predict a data point (infer y given x), we can use the formula:

y = m * x + c + err
based on estimated values of m, c and err.

The uncertainties of the data point is contributed by the uncertainties in m and c, as well as err. In ChartDirector, the uncertainties of the data points are represented visually as a prediction band around the regression line. For example, a 95% prediction band means there are 95% probability that a data point will be in that band.

The prediction band is always wider than the confidence band. It is because the uncertainties of the regression line is contributed by m and c, while the uncertainties of the data points are contributed by m, c and err. The err term makes the data points less certain that the regression line.
Subclass of the CDLayerMBS class.
This is a subclass of an abstract class. You can't create an instance, but you can get one from various plugin functions.

Feedback, Comments & Corrections

Super class CDLayerMBS

This class has no sub classes.

Some methods using this class:

Some examples which use this class:

The items on this page are in the following plugins: MBS ChartDirector Plugin.

CDThreeDScatterGroupMBS   -   CDTTFTextMBS

MBS Xojo PDF Plugins