Trend extraction methods for time series analysis
Batet Cardeñoso, David (2021)
Batet Cardeñoso, David
2021
Matematiikan ja tilastotieteen kandidaattiohjelma - Bachelor's Programme in Mathematics and Statistics
Informaatioteknologian ja viestinnän tiedekunta - Faculty of Information Technology and Communication Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2021-11-30
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-202111158411
https://urn.fi/URN:NBN:fi:tuni-202111158411
Tiivistelmä
The need to extract the trend from a given time series is a common problem in a wide variety of fields, such as signal processing or econometrics. Ordinary least squares regression (OLS), locally weighted polynomial regression (LWP), moving average, wavelet decomposition and empirical mode decomposition (EMD) are examples of methods which are used for trend extraction. In the present work, the aforementioned methods are first presented and then showcased on a simulated time series. Some of the properties of the methods are discussed afterwards.
Kokoelmat
- Kandidaatintutkielmat [7072]