Should we sample a time series more frequently? Decision support via multirate spectrum estimation
Should we sample a time series more frequently? Decision support via multirate spectrum estimation
Suppose we have a historical time series with samples taken at a slow rate, e.g. quarterly. This article proposes a new method to answer the question: is it worth sampling the series at a faster rate, e.g. monthly? Our contention is that classical time series methods are designed to analyse a series at a single and given sampling rate with the consequence that analysts are not often encouraged to think carefully about what an appropriate sampling rate might be. To answer the sampling rate question we propose a novel Bayesian method that incorporates the historical series, cost information and small amounts of pilot data sampled at the faster rate. The heart of our method is a new Bayesian spectral estimation technique that is capable of coherently using data sampled at multiple rates and is demonstrated to have superior practical performance compared to alternatives. Additionally, we introduce a method for hindcasting historical data at the faster rate. A freeware R package, regspec, is available that implements our methods. We illustrate our work using official statistics time series including the United Kingdom consumer price index and counts of United Kingdom residents travelling abroad, but our methods are general and apply to any situation where time series data are collected.
353-407
Nason, Guy P.
1c66079b-871e-47d5-96ed-9e2a859f4ed9
Powell, Ben
866da689-847f-4b75-9150-cd8149684368
Elliott, Duncan
0ff1b380-a7d3-44d7-b7cf-35fb02077d65
Smith, Paul
a2548525-4f99-4baf-a4d0-2b216cce059c
February 2017
Nason, Guy P.
1c66079b-871e-47d5-96ed-9e2a859f4ed9
Powell, Ben
866da689-847f-4b75-9150-cd8149684368
Elliott, Duncan
0ff1b380-a7d3-44d7-b7cf-35fb02077d65
Smith, Paul
a2548525-4f99-4baf-a4d0-2b216cce059c
Nason, Guy P., Powell, Ben, Elliott, Duncan and Smith, Paul
(2017)
Should we sample a time series more frequently? Decision support via multirate spectrum estimation.
Journal of the Royal Statistical Society: Series A (Statistics in Society), 180 (2), .
(doi:10.1111/rssa.12210).
Abstract
Suppose we have a historical time series with samples taken at a slow rate, e.g. quarterly. This article proposes a new method to answer the question: is it worth sampling the series at a faster rate, e.g. monthly? Our contention is that classical time series methods are designed to analyse a series at a single and given sampling rate with the consequence that analysts are not often encouraged to think carefully about what an appropriate sampling rate might be. To answer the sampling rate question we propose a novel Bayesian method that incorporates the historical series, cost information and small amounts of pilot data sampled at the faster rate. The heart of our method is a new Bayesian spectral estimation technique that is capable of coherently using data sampled at multiple rates and is demonstrated to have superior practical performance compared to alternatives. Additionally, we introduce a method for hindcasting historical data at the faster rate. A freeware R package, regspec, is available that implements our methods. We illustrate our work using official statistics time series including the United Kingdom consumer price index and counts of United Kingdom residents travelling abroad, but our methods are general and apply to any situation where time series data are collected.
Text
NPES_R2.pdf
- Accepted Manuscript
More information
Accepted/In Press date: 19 April 2016
e-pub ahead of print date: 18 December 2016
Published date: February 2017
Organisations:
Statistical Sciences Research Institute
Identifiers
Local EPrints ID: 393317
URI: http://eprints.soton.ac.uk/id/eprint/393317
ISSN: 0964-1998
PURE UUID: d5d40905-48e4-428a-a4d6-b1e230c077d7
Catalogue record
Date deposited: 25 Apr 2016 13:57
Last modified: 15 Mar 2024 05:31
Export record
Altmetrics
Contributors
Author:
Guy P. Nason
Author:
Ben Powell
Author:
Duncan Elliott
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics