Predictability limit of partially observed systems
Abstract Applications from finance to epidemiology and cyber-security require accurate forecasts of dynamic phenomena, which are often only partially observed. We demonstrate that a system’s predictability degrades as a function of temporal sampling, regardless of the adopted forecasting model. We q...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2020-11-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-020-77091-1 |
| _version_ | 1819200238625226752 |
|---|---|
| author | Andrés Abeliuk Zhishen Huang Emilio Ferrara Kristina Lerman |
| author_facet | Andrés Abeliuk Zhishen Huang Emilio Ferrara Kristina Lerman |
| author_sort | Andrés Abeliuk |
| collection | DOAJ |
| description | Abstract Applications from finance to epidemiology and cyber-security require accurate forecasts of dynamic phenomena, which are often only partially observed. We demonstrate that a system’s predictability degrades as a function of temporal sampling, regardless of the adopted forecasting model. We quantify the loss of predictability due to sampling, and show that it cannot be recovered by using external signals. We validate the generality of our theoretical findings in real-world partially observed systems representing infectious disease outbreaks, online discussions, and software development projects. On a variety of prediction tasks—forecasting new infections, the popularity of topics in online discussions, or interest in cryptocurrency projects—predictability irrecoverably decays as a function of sampling, unveiling predictability limits in partially observed systems. |
| first_indexed | 2024-12-23T03:29:03Z |
| format | Article |
| id | doaj.art-89a8ba139d4b4b4e9e2e92ddd36a2488 |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-12-23T03:29:03Z |
| publishDate | 2020-11-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-89a8ba139d4b4b4e9e2e92ddd36a24882022-12-21T18:01:45ZengNature PortfolioScientific Reports2045-23222020-11-0110111010.1038/s41598-020-77091-1Predictability limit of partially observed systemsAndrés Abeliuk0Zhishen Huang1Emilio Ferrara2Kristina Lerman3Information Sciences Institute, University of Southern CaliforniaUniversity of Colorado BoulderInformation Sciences Institute, University of Southern CaliforniaInformation Sciences Institute, University of Southern CaliforniaAbstract Applications from finance to epidemiology and cyber-security require accurate forecasts of dynamic phenomena, which are often only partially observed. We demonstrate that a system’s predictability degrades as a function of temporal sampling, regardless of the adopted forecasting model. We quantify the loss of predictability due to sampling, and show that it cannot be recovered by using external signals. We validate the generality of our theoretical findings in real-world partially observed systems representing infectious disease outbreaks, online discussions, and software development projects. On a variety of prediction tasks—forecasting new infections, the popularity of topics in online discussions, or interest in cryptocurrency projects—predictability irrecoverably decays as a function of sampling, unveiling predictability limits in partially observed systems.https://doi.org/10.1038/s41598-020-77091-1 |
| spellingShingle | Andrés Abeliuk Zhishen Huang Emilio Ferrara Kristina Lerman Predictability limit of partially observed systems Scientific Reports |
| title | Predictability limit of partially observed systems |
| title_full | Predictability limit of partially observed systems |
| title_fullStr | Predictability limit of partially observed systems |
| title_full_unstemmed | Predictability limit of partially observed systems |
| title_short | Predictability limit of partially observed systems |
| title_sort | predictability limit of partially observed systems |
| url | https://doi.org/10.1038/s41598-020-77091-1 |
| work_keys_str_mv | AT andresabeliuk predictabilitylimitofpartiallyobservedsystems AT zhishenhuang predictabilitylimitofpartiallyobservedsystems AT emilioferrara predictabilitylimitofpartiallyobservedsystems AT kristinalerman predictabilitylimitofpartiallyobservedsystems |