Seasonal adjustment of high-frequency time series

Date

2023-4

Author

Tosun, İlayda

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

364
views

310
downloads

Cite This

Seasonal adjustment is a statistical methodology employed to eliminate the influence of regular cyclic patterns inherent in time series data. Seasonal adjustment is essential for accurately interpreting and analyzing time series, particularly in macroeconomics, where understanding long-term trends and patterns is crucial for making informed decisions. In recent years, there has been a substantial surge in the availability of high-frequency time series data, which pertains to data collected at very short intervals. Seasonal adjustment of high-frequency time series poses unique challenges due to their tendency to exhibit high levels of noise and volatility. As a result, accurately identifying and removing seasonal effects becomes more challenging. New methods are being developed on this subject. This study aims to assess a diverse range of data sets possessing different frequencies and characteristics, employing innovative approaches. Through the utilization of techniques such as naive, Multivariate Seasonal Trend Decomposition using Loess (MSTL), Seasonal-Trend Decomposition using Regression (STR), and Daily Seasonal Adjustment (DSA), the seasonal adjustment of time series exhibiting intricate seasonal patterns at hourly, daily, and weekly frequencies is undertaken. In particular, when analyzing hourly data, the STR method shows outstanding results, while for daily data, the DSA method performs better based on the four datasets used in the study. In general, both the MSTL and STR methods have shown impressive results.

Subject Keywords

Time series analysis, Seasonal adjustment, High-frequency time series

URI

https://hdl.handle.net/11511/103254

Collections

Graduate School of Applied Mathematics, Thesis