Telescope

Telescope is a hybrid forecasting tool written in R and designed to perform multi-step-ahead forecasts for univariate time series while maintaining a short runtime. The forecasting method is based on STL time series decomposition. To achieve better forecasting results, Telescope uses clustering techniques for categorical information creation and ARIMA, ANN, and XGBoost as forecasting methods. Telescope users can pass a matrix of timestamps and observation values, set the length of the forecasting horizon, and also set various optional parameters.

Download:

Telescope is published under GPL v3 here
FAS Tutorial on "Best practices for Time Series Forecasting" Slides available here

Installation:

This package can be installed in R by using the following commands:

install.packages("devtools")
devtools::install_github("DescartesResearch/telescope")

For unknown reasons, install_gitub does not work under all Windows versions. Therefore the package can alternatively be installed in R with the following commands:

install.packages("remotes")
remotes::install_url(url="https://github.com/DescartesResearch/telescope/archive/master.zip", INSTALL_opt= "--no-multiarch")

Publications

2021[ to top ]

Automatisierte Hybride Zeitreihenprognose A. Bauer; in Ausgezeichnete Informatikdissertationen 2020, S. Hölldobler (Hrsg.) (2021). 29–38.
- [ BibTeX ]
- [ EndNote ]
@inproceedings{mci/Bauer2021, address = {Bonn}, author = {Bauer, André}, booktitle = {Ausgezeichnete Informatikdissertationen 2020}, editor = {Hölldobler, Steffen}, keywords = {telescope}, pages = {29-38}, publisher = {Gesellschaft für Informatik e.V.}, title = {Automatisierte Hybride Zeitreihenprognose}, year = 2021 }

%0 Conference Paper %1 mci/Bauer2021 %A Bauer, André %B Ausgezeichnete Informatikdissertationen 2020 %C Bonn %D 2021 %E Hölldobler, Steffen %I Gesellschaft für Informatik e.V. %P 29-38 %T Automatisierte Hybride Zeitreihenprognose

2020[ to top ]

Time Series Forecasting for Self-Aware Systems A. Bauer; M. Züfle; N. Herbst; A. Zehe; A. Hotho; S. Kounev; in Proceedings of the IEEE (2020). 108(7) 1068–1093.
- [ BibTeX ]
- [ EndNote ]
- [ DOI ]
@article{BaZuHeZeHoKo-PIEEE-Forecasting, author = {Bauer, Andr{\'e} and Z{\"u}fle, Marwin and Herbst, Nikolas and Zehe, Albin and Hotho, Andreas and Kounev, Samuel}, journal = {Proceedings of the IEEE}, keywords = {t_journalmagazine}, month = 7, number = 7, pages = {1068 - 1093}, title = {{Time Series Forecasting for Self-Aware Systems}}, volume = 108, year = 2020 }

%0 Journal Article %1 BaZuHeZeHoKo-PIEEE-Forecasting %A Bauer, André %A Züfle, Marwin %A Herbst, Nikolas %A Zehe, Albin %A Hotho, Andreas %A Kounev, Samuel %D 2020 %J Proceedings of the IEEE %N 7 %P 1068 - 1093 %R 10.1109/JPROC.2020.2983857 %T Time Series Forecasting for Self-Aware Systems %V 108
Telescope: An Automatic Feature Extraction and Transformation Approach for Time Series Forecasting on a Level-Playing Field A. Bauer; M. Züfle; N. Herbst; S. Kounev; V. Curtef; in Proceedings of the 36th International Conference on Data Engineering (ICDE) (2020).
- [ BibTeX ]
- [ EndNote ]
@inproceedings{BaZuHeKoCu-ICDE-Telescope, author = {Bauer, Andr{\'e} and Z{\"u}fle, Marwin and Herbst, Nikolas and Kounev, Samuel and Curtef, Valentin}, booktitle = {Proceedings of the 36th International Conference on Data Engineering (ICDE)}, keywords = {Prediction}, month = {April}, title = {Telescope: An Automatic Feature Extraction and Transformation Approach for Time Series Forecasting on a Level-Playing Field}, year = 2020 }

%0 Conference Paper %1 BaZuHeKoCu-ICDE-Telescope %A Bauer, André %A Züfle, Marwin %A Herbst, Nikolas %A Kounev, Samuel %A Curtef, Valentin %B Proceedings of the 36th International Conference on Data Engineering (ICDE) %D 2020 %T Telescope: An Automatic Feature Extraction and Transformation Approach for Time Series Forecasting on a Level-Playing Field

2019[ to top ]

Challenges and Approaches: Forecasting for Autonomic Computing A. Bauer; in Organic Computing: Doctoral Dissertation Colloquium 2018, S. Tomforde, B. Sick (Hrsg.) (2019).
- [ BibTeX ]
- [ EndNote ]
@incollection{Ba2019-OCDDC-Forecatsing, author = {Bauer, Andr{\'e}}, booktitle = {{Organic Computing: Doctoral Dissertation Colloquium 2018}}, editor = {Tomforde, Sven and Sick, Bernhard}, keywords = {Self-adaptive-systems}, month = {April}, publisher = {kassel university press GmbH}, title = {{Challenges and Approaches: Forecasting for Autonomic Computing}}, year = 2019 }

%0 Book Section %1 Ba2019-OCDDC-Forecatsing %A Bauer, André %B Organic Computing: Doctoral Dissertation Colloquium 2018 %D 2019 %E Tomforde, Sven %E Sick, Bernhard %I kassel university press GmbH %T Challenges and Approaches: Forecasting for Autonomic Computing
Methoden und Messverfahren für Automatisches Skalieren in Elastischen Cloud Umgebungen N. Herbst; (2019, Juni).
- [ BibTeX ]
- [ EndNote ]
@misc{talk-Herbst-GIAwardNomination-Dagstuhl, author = {Herbst, Nikolas}, howpublished = {GI Dissertationspreis: Seminar f{\"u}r Nominierte Doktoranden, Dagstuhl, Germany, June 28, 2019}, keywords = {t_bookchapter}, month = {June}, title = {{Methoden und Messverfahren f{\"u}r Automatisches Skalieren in Elastischen Cloud Umgebungen}}, year = 2019 }

%0 Generic %1 talk-Herbst-GIAwardNomination-Dagstuhl %A Herbst, Nikolas %D 2019 %T Methoden und Messverfahren für Automatisches Skalieren in Elastischen Cloud Umgebungen
Best Practices for Time Series Forecasting A. Bauer; M. Züfle; N. Herbst; S. Kounev; in 2019 IEEE 4th International Workshops on Foundations and Applications of Self* Systems (FAS*W) (2019).
- [ BibTeX ]
- [ EndNote ]
@inproceedings{BaZuHeKo2019-ICAC-ForecastingTutorial, author = {Bauer, Andr{\'e} and Z{\"u}fle, Marwin and Herbst, Nikolas and Kounev, Samuel}, booktitle = {{2019 IEEE 4th International Workshops on Foundations and Applications of Self* Systems (FAS*W)}}, keywords = {t_tutorial}, month = {June}, title = {{Best Practices for Time Series Forecasting}}, year = 2019 }

%0 Conference Paper %1 BaZuHeKo2019-ICAC-ForecastingTutorial %A Bauer, André %A Züfle, Marwin %A Herbst, Nikolas %A Kounev, Samuel %B 2019 IEEE 4th International Workshops on Foundations and Applications of Self* Systems (FAS*W) %D 2019 %T Best Practices for Time Series Forecasting

2018[ to top ]

Comparing Machine Learning Approaches for Multivariate Time Series Forecasting D. Otto; Thesis; University of Würzburg; Am Hubland, Informatikgebäude, 97074 Würzburg, Germany. (2018, April).
- [ BibTeX ]
- [ EndNote ]
@mastersthesis{OttoMultiTelescope2018, address = {Am Hubland, Informatikgeb{\"a}ude, 97074 W{\"u}rzburg, Germany}, author = {Otto, Dominik}, keywords = {Optimization}, month = {April}, school = {University of W{\"u}rzburg}, title = {{Comparing Machine Learning Approaches for Multivariate Time Series Forecasting}}, type = {{Bachelor Thesis}}, year = 2018 }

%0 Thesis %1 OttoMultiTelescope2018 %A Otto, Dominik %C Am Hubland, Informatikgebäude, 97074 Würzburg, Germany %D 2018 %T Comparing Machine Learning Approaches for Multivariate Time Series Forecasting
Automatisiertes Forecasting mit Feature Engineering B. Augustin; Thesis; University of Würzburg; Am Hubland, Informatikgebäude, 97074 Würzburg, Germany. (2018, April).
- [ BibTeX ]
- [ EndNote ]
@mastersthesis{BrittaAFFE2018, address = {Am Hubland, Informatikgeb{\"a}ude, 97074 W{\"u}rzburg, Germany}, author = {Augustin, Britta}, keywords = {Automated_model_learning}, month = {April}, school = {University of W{\"u}rzburg}, title = {{Automatisiertes Forecasting mit Feature Engineering}}, type = {{Bachelor Thesis}}, year = 2018 }

%0 Thesis %1 BrittaAFFE2018 %A Augustin, Britta %C Am Hubland, Informatikgebäude, 97074 Würzburg, Germany %D 2018 %T Automatisiertes Forecasting mit Feature Engineering
Predictive Maintenance for Industry 4.0 M. Züfle; (2018, April).
- [ BibTeX ]
- [ EndNote ]
@misc{talk-Zuefle-SelPhyS, author = {Z{\"u}fle, Marwin}, howpublished = {Talk at Workshop on Self-Awareness in Cyber-Physical Systems 2018}, keywords = {Talks}, month = {April}, title = {{Predictive Maintenance for Industry 4.0}}, year = 2018 }

%0 Generic %1 talk-Zuefle-SelPhyS %A Züfle, Marwin %D 2018 %T Predictive Maintenance for Industry 4.0
Methods and Benchmarks for Auto-Scaling Mechanisms in Elastic Cloud Environments N. Herbst; Thesis; University of Würzburg, Germany. (2018, Juli).
- [ BibTeX ]
- [ EndNote ]
- [ URL ]
@phdthesis{Herbst2018-phd, author = {Herbst, Nikolas}, keywords = {SPEC}, month = {July}, note = {<b>{Distinction, SPEC Kaivalya Dixit Distinguished Dissertation Award 2018}</b>}, school = {University of W{\"u}rzburg, Germany}, title = {Methods and Benchmarks for Auto-Scaling Mechanisms in Elastic Cloud Environments}, year = 2018 }

%0 Thesis %1 Herbst2018-phd %A Herbst, Nikolas %D 2018 %T Methods and Benchmarks for Auto-Scaling Mechanisms in Elastic Cloud Environments %U https://opus.bibliothek.uni-wuerzburg.de/frontdoor/index/index/docId/16431
How we built a scalable micro-service application - The Tea Store N. Herbst; (2018, Juli).
- [ BibTeX ]
- [ EndNote ]
- [ URL ]
@misc{talk-Herbst-ScrumScale, author = {Herbst, Nikolas}, howpublished = {Keynote Talk at the ScrumScale workshop, Oslo, Norway, July 5, 2018}, keywords = {SPEC}, month = {July}, title = {{How we built a scalable micro-service application - The Tea Store}}, year = 2018 }

%0 Generic %1 talk-Herbst-ScrumScale %A Herbst, Nikolas %D 2018 %T How we built a scalable micro-service application - The Tea Store %U https://www.dataforeningen.no/smidig-skalerbarhet.6113996-134138.html

2017[ to top ]

Telescope: A Hybrid Forecast Method for Univariate Time Series M. Züfle; A. Bauer; N. Herbst; V. Curtef; S. Kounev; in Proceedings of the International work-conference on Time Series (ITISE 2017) (2017).
- [ Abstract ]
- [ BibTeX ]
- [ EndNote ]
Forecasting is an important part of the decision-making process and used in many fields like business, economics, finance, science, and engineering. According to the No-Free-Lunch-Theorem from 1997, there is no general forecasting method, that performs best for all time series. Instead, expert knowledge is needed to decide which forecasting method to choose for a specific time series with its own characteristics. Since a trial and error approach is very inefficient and expert knowledge is useful but a time-consuming task that cannot be fully automated, we present a new hybrid multi-step-ahead forecasting approach based on time series decomposition. Initial evaluations show that this hybrid approach improves the forecast accuracy compared to six existing forecasting methods while maintaining a short runtime.

@inproceedings{ZBBHCK2017-ITISE-Telescope, abstract = {Forecasting is an important part of the decision-making process and used in many fields like business, economics, finance, science, and engineering. According to the No-Free-Lunch-Theorem from 1997, there is no general forecasting method, that performs best for all time series. Instead, expert knowledge is needed to decide which forecasting method to choose for a specific time series with its own characteristics. Since a trial and error approach is very inefficient and expert knowledge is useful but a time-consuming task that cannot be fully automated, we present a new hybrid multi-step-ahead forecasting approach based on time series decomposition. Initial evaluations show that this hybrid approach improves the forecast accuracy compared to six existing forecasting methods while maintaining a short runtime.}, author = {Z{\"u}fle, Marwin and Bauer, Andr{\'e} and Herbst, Nikolas and Curtef, Valentin and Kounev, Samuel}, booktitle = {{Proceedings of the International work-conference on Time Series (ITISE 2017)}}, keywords = {Optimization}, month = {September}, title = {{Telescope: A Hybrid Forecast Method for Univariate Time Series}}, year = 2017 }

%0 Conference Paper %1 ZBBHCK2017-ITISE-Telescope %A Züfle, Marwin %A Bauer, André %A Herbst, Nikolas %A Curtef, Valentin %A Kounev, Samuel %B Proceedings of the International work-conference on Time Series (ITISE 2017) %D 2017 %T Telescope: A Hybrid Forecast Method for Univariate Time Series %X Forecasting is an important part of the decision-making process and used in many fields like business, economics, finance, science, and engineering. According to the No-Free-Lunch-Theorem from 1997, there is no general forecasting method, that performs best for all time series. Instead, expert knowledge is needed to decide which forecasting method to choose for a specific time series with its own characteristics. Since a trial and error approach is very inefficient and expert knowledge is useful but a time-consuming task that cannot be fully automated, we present a new hybrid multi-step-ahead forecasting approach based on time series decomposition. Initial evaluations show that this hybrid approach improves the forecast accuracy compared to six existing forecasting methods while maintaining a short runtime.

Telescope

Telescope

Download:

Installation:

Publications

Mailing List

Hinweis zum Datenschutz

Hinweis zum Datenschutz

Bildnachweise

Your E-mail address:
Your Name (optional):