Detecting Distributional Differences between Temporal Granularities for Exploratory Time Series Analysis

Sayani Gupta (), Rob Hyndman and Dianne Cook ()

No 20/21, Monash Econometrics and Business Statistics Working Papers from Monash University, Department of Econometrics and Business Statistics

Abstract: Cyclic temporal granularities are temporal deconstructions of a time period into units such as hour-of-theday and work-day/weekend. They can be useful for measuring repetitive patterns in large univariate time series data, and feed new approaches to exploring time series data. One use is to take pairs of granularities, and make plots of response values across the categories induced by the temporal deconstruction. However, when there are many granularities that can be constructed for a time period, there will also be too many possible displays to decide which might be the more interesting to display. This work proposes a new distance metric to screen and rank the possible granularities, and hence choose the most interesting ones to plot. The distance measure is computed for a single or pairs of cyclic granularities and can be compared across different cyclic granularities or on a collection of time series. The methods are implemented in the open-source R package hakear.

Keywords: data visualization; cyclic granularities; periodicities; permutation tests; distributional difference; Jensen-Shannon distances; smart meter data; R (search for similar items in EconPapers)
JEL-codes: C55 C65 C80 (search for similar items in EconPapers)
Pages: 24
Date: 2021
New Economics Papers: this item is included in nep-ecm, nep-ets and nep-ore
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.monash.edu/business/ebs/research/publications/ebs/wp20-2021.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:msh:ebswps:2021-20

Ordering information: This working paper can be ordered from
http://business.mona ... -business-statistics

Access Statistics for this paper

More papers in Monash Econometrics and Business Statistics Working Papers from Monash University, Department of Econometrics and Business Statistics PO Box 11E, Monash University, Victoria 3800, Australia. Contact information at EDIRC.
Bibliographic data for series maintained by Professor Xibin Zhang ().