1,683
edits
Changes
Jump to navigation
Jump to search
mLine 79:
Line 79: − + +
Finding relations between variables in time series (edit)
Revision as of 22:39, 3 July 2023
, 22:39, 3 July 2023no edit summary
Types of data. [Exercised] is an event with specific occurrence moment and length while [tired] is a vaguer value user could use to try to describe feelings past 4 hours.
Types of data. [Exercised] is an event with specific occurrence moment and length while [tired] is a vaguer value user could use to try to describe feelings past 4 hours.
=== What to expect from the complete analysis tool ===
The difference between an app finding a relationship between two variables through correct and incorrect means maybe very difficult to detect. However, the graphs it produces should include most of the following: Interpolation for irregular time series. Change point or breakpoint detection. Outlier detection. Smoothing. Cycles decomposition using a model like ARIMA. Ex. kayak season is in the summer or lunch is at exactly 1pm. Not consistently timed event shape detection like dinner is anywhere between 4 and 10pm and causes a particular 2 hour spike in glucose (I know this is a bad example but ...). Removal of effect of variables found to correlate with this one to show residuals.
[[Category:Data analysis]]
[[Category:Data analysis]]