# Surrogate data

> Mediated Wiki article. Canonical URL: https://mediated.wiki/source/Surrogate_data
> Markdown URL: https://mediated.wiki/source/Surrogate_data.md
> Source: https://en.wikipedia.org/wiki/Surrogate_data
> Source revision: 1321451684
> License: Creative Commons Attribution-ShareAlike 4.0 International (https://creativecommons.org/licenses/by-sa/4.0/)

{{Short description|Time series data technique}}
'''Surrogate data''', sometimes known as '''analogous data''',<ref name=Kaefer /> usually refers to [time series](/source/time_series) data that is produced using well-defined (linear) models like [ARMA](/source/Autoregressive%E2%80%93moving-average_model) processes that reproduce various statistical properties like the [autocorrelation](/source/autocorrelation) structure of a measured data set.<ref name=Prichard>{{cite journal
 |title=Generating surrogate data for time series with several simultaneously measured variables
 |url=http://public.lanl.gov/jt/Papers/SurrogateMultiple.pdf
 |journal=Physical Review Letters
 |volume=73
 |number=7
 |pages=951–954
 |year=1994
 |author1=Prichard
 |author2=Theiler
 |doi=10.1103/physrevlett.73.951
 |pmid=10057582|arxiv=comp-gas/9405002
 |bibcode=1994PhRvL..73..951P
 |s2cid=32748996
 }}</ref> The resulting surrogate data can then for example be used for testing for non-linear structure in the empirical data; this is called [surrogate data testing](/source/surrogate_data_testing).

Surrogate or analogous data also refers to data used to supplement available data from which a [mathematical model](/source/mathematical_model) is built. Under this definition, it may be generated (i.e., [synthetic data](/source/synthetic_data)) or transformed from another source.<ref name=Kaefer>{{cite thesis |degree=M.Sc. |last=Kaefer |first=Paul E. |date=2015 |title=Transforming Analogous Time Series Data to Improve Natural Gas Demand Forecast Accuracy |publisher=Marquette University |url=http://epublications.marquette.edu/theses_open/320/ |access-date=2016-02-18 |archive-date=2016-03-12 |archive-url=https://web.archive.org/web/20160312044649/http://epublications.marquette.edu/theses_open/320/ |url-status=live }}</ref>

== Uses ==
Surrogate data is used in environmental and laboratory settings, when study data from one source is used in estimation of characteristics of another source.<ref>{{cite web |url=http://www.caslab.com/Surrogate_Data_Meaning/ |title=Surrogate Data Meaning |publisher=Columbia Analytical Services, Inc., now ALS Environmental |quote=What is Surrogate Data? Data from studies of test organisms or a test substance that are used to estimate the characteristics or effects on another organism or substance. |access-date=February 15, 2017 |archive-date=February 16, 2017 |archive-url=https://web.archive.org/web/20170216130237/http://www.caslab.com/Surrogate_Data_Meaning/ |url-status=live }}</ref> For example, it has been used to model population trends in animal species.<ref name=PLOSONE>{{cite journal |title=The Use of Surrogate Data in Demographic Population Viability Analysis: A Case Study of California Sea Lions |first1=Claudia J. |last1=Hernández-Camacho |first2=Victoria. J. |last2=Bakker |first3=David |last3=Aurioles-Gamboa |first4=Jeff |last4=Laake |first5=Leah R. |last5=Gerber |author-link5=Leah Gerber|editor=Aaron W. Reed |journal=[PLOS ONE](/source/PLOS_ONE) |volume=10 |number=9 |date=September 2015 |doi=10.1371/journal.pone.0139158 |pmid=26413746 |article-number=e0139158|bibcode=2015PLoSO..1039158H |pmc=4587556 |doi-access=free }}</ref> It can also be used to model biodiversity, as it would be difficult to gather actual data on all species in a given area.<ref name=Biodiversity>{{cite journal |title=Environmental diversity: on the best-possible use of surrogate data for assessing the relative biodiversity of sets of areas |first1=D.P. |last1=Faith |first2=P.A. |last2=Walker |journal=Biodiversity and Conservation |volume=5 |number=4 |year=1996 |publisher=Springer Nature |pages=399–415 |doi=10.1007/BF00056387|bibcode=1996BiCon...5..399F |s2cid=24066193 }}</ref>

Surrogate data may be used in forecasting. Data from similar series may be pooled to improve forecast accuracy.<ref>{{cite book |chapter=Forecasting Analogous Time Series |title=Principles of Forecasting: A Handbook for Researchers and Practitioners |first1=George T. |last1=Duncan |first2=Wilpen L. |last2=Gorr |first3=Janusz |last3=Szczypula |editor=J. Scott Armstrong |editor-link=J. Scott Armstrong |publisher=Kluwer Academic Publishers |year=2001 |pages=195–213 |isbn=0-7923-7930-6}}</ref> Use of surrogate data may enable a model to account for patterns not seen in historical data.<ref>{{cite conference |title=Using Surrogate Data to Mitigate the Risks of Natural Gas Forecasting on Unusual Days |url=https://forecasters.org/wp-content/uploads/gravity_forms/7-621289a708af3e7af65a7cd487aee6eb/2015/07/Kaefer_Paul_ISF2015.pdf |first1=Paul E. |last1=Kaefer |first2=Babatunde |last2=Ishola |first3=Ronald H. |last3=Brown |first4=George F. |last4=Corliss |conference=International Institute of Forecasters: 35th International Symposium on Forecasting |year=2015 |website=forecasters.org/isf |access-date=2022-07-20 |archive-date=2021-05-17 |archive-url=https://web.archive.org/web/20210517041956/https://forecasters.org/wp-content/uploads/gravity_forms/7-621289a708af3e7af65a7cd487aee6eb/2015/07/Kaefer_Paul_ISF2015.pdf |url-status=live }}</ref>

Another use of surrogate data is to test models for non-linearity. The term [surrogate data testing](/source/surrogate_data_testing) refers to algorithms used to analyze models in this way.<ref name=Schreiber>{{cite journal |first1=Thomas |last1=Schreiber |first2=Andreas |last2=Schmitz |title=Surrogate time series |journal=Physica D |volume=142 |issue=3–4 |pages=346–382 |year=1999 |doi=10.1016/s0167-2789(00)00043-9|arxiv=chao-dyn/9909037|bibcode=2000PhyD..142..346S |citeseerx=10.1.1.46.3999 |s2cid=13889229 }}</ref> These tests typically involve generating data, whereas surrogate data in general can be produced or gathered in many ways.<ref name=Kaefer />

== Methods ==
One method of surrogate data is to find a source with similar conditions or parameters, and use those data in modeling.<ref name=PLOSONE /> Another method is to focus on patterns of the underlying system, and to search for a similar pattern in related data sources (for example, patterns in other related species or environmental areas).<ref name=Biodiversity />

Rather than using existing data from a separate source, surrogate data may be generated through statistical processes,<ref name=Prichard /> which may involve random data generation<ref name=Kaefer /> using constraints of the model or system.<ref name=Schreiber />

== See also ==
* [Bootstrapping (statistics)](/source/Bootstrapping_(statistics))
* [Data augmentation](/source/Data_augmentation)
* [Jackknife resampling](/source/Jackknife_resampling)
* [Synthetic data](/source/Synthetic_data)

==References==
{{Reflist|30em}}

==Further reading==
* {{Cite journal | last1 = Schreiber | first1 = T. | last2 = Schmitz | first2 = A. | doi = 10.1103/PhysRevLett.77.635 | title = Improved Surrogate Data for Nonlinearity Tests | journal = Physical Review Letters | volume = 77 | issue = 4 | pages = 635–638 | year = 1996 | pmid =  10062864|bibcode = 1996PhRvL..77..635S | arxiv = chao-dyn/9909041 | s2cid = 13193081 }}

Category:Statistical data types
Category:Nonlinear time series analysis

---
Adapted from the Wikipedia article [Surrogate data](https://en.wikipedia.org/wiki/Surrogate_data) by Wikipedia contributors ([contributor history](https://en.wikipedia.org/wiki/Surrogate_data?action=history)). Available under [Creative Commons Attribution-ShareAlike 4.0 International](https://creativecommons.org/licenses/by-sa/4.0/). Changes may have been made.
