Cumulative Sums and Visual Change Detection between Two Random Processes

Requires a Wolfram Notebook System

Interact on desktop, mobile and cloud with the free Wolfram CDF Player or other Wolfram Language products.

Requires a Wolfram Notebook System

Edit on desktop, mobile and cloud with any Wolfram Language product.

This Demonstration illustrates a simplified version of the method of cumulative sums (CUSUM). A dataset of the form , , …, is constructed by drawing the first observations at random from a normal distribution with mean specified by the "left process" control and the remainder from a normal distribution with mean specified by the "right process" control. The standard deviation in both cases is 3. The top graph shows the dataset, while the bottom graph shows the cumulative sums of the dataset (i.e., the values , , …, where ). It is arguably easier to detect the boundary between the two processes from the second plot than from the first.

Contributed by: Olexandr Eugene Prokopchenko and Pylyp Prokopchenko (September 2012)
Open content licensed under CC BY-NC-SA


Snapshots


Details

detailSectionParagraph


Feedback (field required)
Email (field required) Name
Occupation Organization
Note: Your message & contact information may be shared with the author of any specific Demonstration for which you give feedback.
Send