scikit-forecasts: Una librería en Python para el pronóstico de series de tiempo no lineales
dc.contributor.advisor | Velásquez Henao, Juan David | |
dc.contributor.author | Arango González, María Alejandra | |
dc.date.accessioned | 2022-08-19T15:56:45Z | |
dc.date.available | 2022-08-19T15:56:45Z | |
dc.date.issued | 2021 | |
dc.description | ilustraciones, diagramas, tablas | spa |
dc.description.abstract | El interés en el uso de técnicas de aprendizaje de máquinas, y en general, de modelos no lineales para el pronóstico de series de tiempo ha crecido exponencialmente en las últimas dos décadas. Sin embargo, muchas de las librerías disponibles de aprendizaje de máquinas no contemplan el uso de modelos de series de tiempo, haciendo que el científico de datos consuma gran parte de su tiempo en la tarea de convertir sus datos a un formato de problema de regresión para poder utilizar estas librerías. Esto claramente evidencia la necesidad de contar con librerías especializadas en el pronóstico de series de tiempo usando técnicas de aprendizaje de máquinas y modelos no lineales en general. En esta tesis se presenta la librería de Python de código abierto llamada scikit-forecasts, la cual permite transformar y pronosticar series de tiempo usando técnicas de aprendizaje de máquinas, entre las que se incluyen los modelos autorregresivos, las redes neuronales artificiales, modelos neuro-difusos, modelos TAR y modelos SETAR, entre otros. La librería puede ser usada interactivamente en un libro de Jupyter, lo que facilita el desarrollo de modelos. Los métodos de pronóstico se encuentran implementados como clases que pueden ser utilizadas con muchas de las funciones disponibles en scikit-learn. La librería está diseñada para soportar los procesos de evaluación de distintos tipos de transformaciones, el pronóstico con diferentes tipos de modelos no lineales y la comparación de pronósticos obtenidos con modelos diferentes. (Texto tomado de la fuente) | spa |
dc.description.abstract | The significance of using machine learning techniques and nonlinear models for time series forecasting has been grown exponentially in past two decades. Nevertheless, a lot of Python packages and libraries using machine learning methods are not taking in account time series models and making data scientists to consume a lot of their time converting data into a regression problem format in order to use these libraries. This kind of issue remarks a necessity of packages and libraries focused on time series forecasting and the use of machine learning techniques and nonlinear models in general. This thesis introduces an open-source Python package called scikit-forecasts, which allows time series transforming and forecasting by using machine learning techniques such as autoregressive modes, artificial neural networks, neuro-fuzzy models, TAR and SETAR models, among others. This package can be interactively used in Jupyter notebooks, which simplifies and promotes new models’ development. Forecasting methods are implemented as classes that can be used with scikit-learn’s available functions. This package has been designed to support evaluation processes of different types of transformations, forecasts of several types of nonlinear models and comparisons of forecasting results that have been obtained using different models. | eng |
dc.description.curriculararea | Área Curricular de Ingeniería de Sistemas e Informática | spa |
dc.description.degreelevel | Maestría | spa |
dc.description.degreename | Magister en Ingeniería de Sistemas | spa |
dc.description.researcharea | Analítica | spa |
dc.format.extent | xi, 65 páginas | spa |
dc.format.mimetype | application/pdf | spa |
dc.identifier.instname | Universidad Nacional de Colombia | spa |
dc.identifier.reponame | Repositorio Institucional Universidad Nacional de Colombia | spa |
dc.identifier.repourl | https://repositorio.unal.edu.co/ | spa |
dc.identifier.uri | https://repositorio.unal.edu.co/handle/unal/81979 | |
dc.language.iso | spa | spa |
dc.publisher | Universidad Nacional de Colombia | spa |
dc.publisher.branch | Universidad Nacional de Colombia - Sede Medellín | spa |
dc.publisher.department | Departamento de la Computación y la Decisión | spa |
dc.publisher.faculty | Facultad de Minas | spa |
dc.publisher.place | Medellín, Colombia | spa |
dc.publisher.program | Medellín - Minas - Maestría en Ingeniería - Ingeniería de Sistemas | spa |
dc.relation.references | D. R. Brillinger, “Time Series: General,” 2000. | spa |
dc.relation.references | R. S. Tsay, “Time series and forecasting: Brief history and future research,” in Statistics in the 21st Century, CRC Press, 2001, pp. 121–131. doi: 10.2307/2669408 | spa |
dc.relation.references | F. Perez, B. E. Granger, U. C. Berkeley, C. Poly, and S. L. Obispo, “AN OPEN SOURCE FRAMEWORK FOR INTERACTIVE, COLLABORATIVE AND REPRODUCIBLE SCIENTIFIC COMPUTING AND EDUCATION.” | spa |
dc.relation.references | C. M. Kelty, Two bits the cultural significance of free software. 2008 | spa |
dc.relation.references | “Top Programming Languages 2020 - IEEE Spectrum.” https://spectrum.ieee.org/at-work/tech-careers/top-programming-language-2020?referrer=%2F (accessed Apr. 20, 2021). | spa |
dc.relation.references | “Series de Tiempo : Pricing.” https://www.pricing.cl/conocimiento/series-de-tiempo/ (accessed Aug. 18, 2021). | spa |
dc.relation.references | “The 2017 Top Programming Languages - IEEE Spectrum.” https://spectrum.ieee.org/computing/software/the-2017-top-programming-languages (accessed Apr. 20, 2021). | spa |
dc.relation.references | “pandas - Python Data Analysis Library.” https://pandas.pydata.org/ (accessed Apr. 20, 2021). | spa |
dc.relation.references | “Matlab CAPTAIN Toolbox for time series analysis and forecasting.” http://www.es.lancs.ac.uk/cres/captain/default.htm (accessed Apr. 20, 2021). | spa |
dc.relation.references | “Deep Learning Toolbox - MATLAB.” https://www.mathworks.com/products/deep-learning.html (accessed Apr. 20, 2021). | spa |
dc.relation.references | “Package ‘forecast’ Title Forecasting Functions for Time Series and Linear Models Description Methods and tools for displaying and analysing univariate time series forecasts including exponential smoothing via state space models and automatic ARIMA modelling,” 2021 | spa |
dc.relation.references | “Package ‘neuralnet’ Title Training of Neural Networks,” 2019. | spa |
dc.relation.references | “SAS/ETS Time Series Forecasting System.” https://support.sas.com/rnd/app/ets/cap/ets_forecasting.html (accessed Apr. 20, 2021). | spa |
dc.relation.references | “Title stata.com forecast-Econometric model forecasting.” | spa |
dc.relation.references | “SPSS Software | IBM.” https://www.ibm.com/analytics/spss-statistics-software (accessed Apr. 20, 2021). | spa |
dc.relation.references | “Introduction — statsmodels.” https://www.statsmodels.org/stable/index.html (accessed Apr. 20, 2021). | spa |
dc.relation.references | D. van Kuppevelt, C. Meijer, F. Huber, A. van der Ploeg, S. Georgievska, and V. T. van Hees, “Mcfly: Automated deep learning on time series,” SoftwareX, vol. 12, Jul. 2020, doi: 10.1016/j.softx.2020.100548. | spa |
dc.relation.references | “Welcome to TSFEL documentation! — TSFEL 0.1.4 documentation.” https://tsfel.readthedocs.io/en/latest/ (accessed Apr. 20, 2021). | spa |
dc.relation.references | “Software de hojas de cálculo Microsoft Excel | Microsoft 365.” https://www.microsoft.com/es-es/microsoft-365/excel (accessed Apr. 20, 2021). | spa |
dc.relation.references | “NeuralTools para análisis de predicción usando redes neuronales inteligentes - Palisade.” https://www.palisade-lta.com/neuraltools/ (accessed Jul. 02, 2021). | spa |
dc.relation.references | Time Series Analysis and Its Applications - With R Examples | Robert H. Shumway | Springer. 2006. | spa |
dc.relation.references | “Nonlinear Time Series Models 18.1 Introduction.” | spa |
dc.relation.references | R. J. Hyndman and G. Athanasopoulus, “Forecasting: Principles and Practice (2nd ed).” https://otexts.com/fpp2/ (accessed Apr. 20, 2021). | spa |
dc.relation.references | “TensorFlow.” https://www.tensorflow.org/?hl=es-419 (accessed Apr. 20, 2021). | spa |
dc.relation.references | “Keras: the Python deep learning API.” https://keras.io/ (accessed Apr. 20, 2021). | spa |
dc.relation.references | “PyTorch.” https://pytorch.org/ (accessed Apr. 20, 2021). | spa |
dc.relation.references | “scikit-learn: machine learning in Python — scikit-learn 0.24.1 documentation.” https://scikit-learn.org/stable/ (accessed Apr. 20, 2021). | spa |
dc.relation.references | “Time Series Analysis: Forecasting and Control - George E. P. Box, Gwilym M. Jenkins, Gregory C. Reinsel, Greta M. Ljung - Google Libros.” https://books.google.com.co/books?hl=es&lr=&id=rNt5CgAAQBAJ&oi=fnd&pg=PR7&dq=Time+Series+Analysis,+Forecasting+and+Control.+box+jenkins&ots=DK28BOj_QG&sig=HkqPSrBFGGDao_IztPkTUtXxxH0#v=onepage&q=Time Series Analysis%2C Forecasting and Control. box jenkins&f=false (accessed Apr. 30, 2021). | spa |
dc.relation.references | M. G. Bulmer, “A Statistical Analysis of the 10-Year Cycle in Canada,” The Journal of Animal Ecology, vol. 43, no. 3, p. 701, Oct. 1974, doi: 10.2307/3532. | spa |
dc.relation.references | “SILSO | World Data Center for the production, preservation and dissemination of the international sunspot number.” http://sidc.oma.be/silso/home (accessed Apr. 30, 2021). | spa |
dc.relation.references | “UCI Machine Learning Repository: Internet Usage Data Data Set.” https://archive.ics.uci.edu/ml/datasets/Internet+Usage+Data (accessed Apr. 30, 2021). | spa |
dc.relation.references | C. Beaumont, S. Makridakis, S. C. Wheelwright, and V. E. McGee, “Forecasting: Methods and Applications,” J Oper Res Soc, vol. 35, no. 1, p. 79, Jan. 1984, doi: 10.2307/2581936. | spa |
dc.relation.references | J. D. V. HENAO, C. O. Z. PEREZ, and C. J. F. CARDONA, “A COMPARISON OF EXPONENTIAL SMOOTHING AND NEURAL NETWORKS IN TIME SERIES PREDICTION,” DYNA, vol. 80, no. 182, pp. 66–73, Nov. 2013. | spa |
dc.relation.references | “Time Series Forecast Study with Python: Monthly Sales of French Champagne.” https://machinelearningmastery.com/time-series-forecast-study-python-monthly-sales-french-champagne/ (accessed Apr. 20, 2021). | spa |
dc.relation.references | “RPubs - Análisis de normalidad.” https://rpubs.com/PAVelasquezVasconez/354989 (accessed Apr. 20, 2021). | spa |
dc.relation.references | C. M. Jarque and A. K. Bera, “Efficient tests for normality, homoscedasticity and serial independence of regression residuals,” Economics Letters, vol. 6, no. 3, pp. 255–259, 1980, doi: 10.1016/0165-1765(80)90024-5. | spa |
dc.relation.references | J. D. Velásquez, “Adaptive Multidimensional Neuro-Fuzzy Inference System for Time Series Prediction,” IEEE Latin America Transactions, vol. 13, no. 8, pp. 2694–2699, Aug. 2015, doi: 10.1109/TLA.2015.7332151. | spa |
dc.relation.references | M. Ghiassi, H. Saidane, and D. K. Zimbra, “A dynamic artificial neural network model for forecasting time series events,” International Journal of Forecasting, vol. 21, no. 2, pp. 341–362, Apr. 2005, doi: 10.1016/j.ijforecast.2004.10.008. | spa |
dc.relation.references | R. S. Tsay, “Analysis of Financial Time Series Financial Econometrics.” | spa |
dc.rights.accessrights | info:eu-repo/semantics/openAccess | spa |
dc.rights.license | Atribución-NoComercial 4.0 Internacional | spa |
dc.rights.uri | http://creativecommons.org/licenses/by-nc/4.0/ | spa |
dc.subject.ddc | 000 - Ciencias de la computación, información y obras generales | spa |
dc.subject.lemb | Aprendizaje automático (Inteligencia artificial) | |
dc.subject.lemb | Machine learning | |
dc.subject.proposal | predicción | spa |
dc.subject.proposal | series de tiempo | spa |
dc.subject.proposal | redes neuronales artificiales | spa |
dc.subject.proposal | no linealidad | spa |
dc.subject.proposal | aprendizaje de máquinas | spa |
dc.title | scikit-forecasts: Una librería en Python para el pronóstico de series de tiempo no lineales | spa |
dc.title.translated | scikit-forecasts: A Python package for nonlinear time series forecasting | eng |
dc.type | Trabajo de grado - Maestría | spa |
dc.type.coar | http://purl.org/coar/resource_type/c_bdcc | spa |
dc.type.coarversion | http://purl.org/coar/version/c_ab4af688f83e57aa | spa |
dc.type.content | Text | spa |
dc.type.driver | info:eu-repo/semantics/masterThesis | spa |
dc.type.redcol | http://purl.org/redcol/resource_type/TM | spa |
dc.type.version | info:eu-repo/semantics/acceptedVersion | spa |
dcterms.audience.professionaldevelopment | Estudiantes | spa |
dcterms.audience.professionaldevelopment | Investigadores | spa |
oaire.accessrights | http://purl.org/coar/access_right/c_abf2 | spa |
Archivos
Bloque original
1 - 1 de 1
Cargando...
- Nombre:
- 1152440067.2021.pdf
- Tamaño:
- 1.09 MB
- Formato:
- Adobe Portable Document Format
- Descripción:
- Tesis de Maestría en Ingeniería de Sistemas
Bloque de licencias
1 - 1 de 1
No hay miniatura disponible
- Nombre:
- license.txt
- Tamaño:
- 3.98 KB
- Formato:
- Item-specific license agreed upon to submission
- Descripción: