find and test break-points

Finding and testing break-points in conventionally formatted air quality data sets.

Usage

findBreakPoints(data, pollutant, h = 0.15, ...)

testBreakPoints(data, pollutant, breaks, ...)

Arguments

data

Data source, typically a data.frame or similar, containing data-series to apply function to and a paired time-stamped data-series, called date.

pollutant

Name of time-series, assumed to be a column in date.

h

(findBreakPoints only) The data/time window size to use when looking for breaks in a supplied time-series, expressed as proportion of time-series (0-1), default 0.15.

...

other parameters

breaks

(testBreakPoints only) data.frame of The break-points and confidence intervals, typically a findBreakPoints output.

Value

findBreakPoints returns a data.frame

of found break-points.

testBreakPoints return a likely break-point/segment report.

Details

findBreakPoints uses methods from strucchange package (see references) and modifications as suggested by the main author of strucchange to handle missing cases to find potential breaks-points in a supplied time-series.

testBreakPoints tests and identifies most likely break-points using methods proposed for use with quantBreakPoints and quantBreakSegments and conventionally formatted air quality data sets.

References

Regarding strucchange methods see breakpoints, and:

Achim Zeileis, Friedrich Leisch, Kurt Hornik and Christian Kleiber (2002). strucchange: An R Package for Testing for Structural Change in Linear Regression Models. Journal of Statistical Software, 7(2), 1-38. URL https://www.jstatsoft.org/v07/i02/.

Achim Zeileis, Christian Kleiber, Walter Kraemer and Kurt Hornik (2003). Testing and Dating of Structural Changes in Practice. Computational Statistics & Data Analysis, 44, 109-123.

Regarding missing data handling, see:

URL: https://stackoverflow.com/questions/43243548/strucchange-not-reporting-breakdates.

Regarding testBreakPoints, see:

Ropkins et al (In Prep).

See also

find.breaks.