TY - GEN

T1 - A Method for Visualizing the Dynamic Characteristics of Economic Time Series Data

AU - Гадасина, Людмила Викторовна

AU - Лабуткин, Иван Алексеевич

N1 - Gadasina, L., Labutkin, I. (2026). A Method for Visualizing the Dynamic Characteristics of Economic Time Series Data. In: Jallouli, R., Bach Tobji, M.A., Omrani, N., Jenhani, I. (eds) Digital Economy. Emerging Technologies and Business Innovation. ICDEc 2025. Lecture Notes in Business Information Processing, vol 560. Springer, Cham.

PY - 2026

Y1 - 2026

N2 - When analysing short-term economic and financial data, it is important to consider phenomena such as economic cycles, trends, and their changes, as well as structural shifts and financial bubbles. This study proposes an effective tool for visualizing these phenomena for high-frequency data, which refers to data collected on a weekly, daily, or more frequent basis. The method involves several steps. First, the choice of a time interval, the change within which is insignificant for analysis purposes. Next, the time series is smoothed using a filter that ignores these intervals. Then, the increments of the time series are calculated by taking differences between neighbouring values. Finally, a phase portrait is constructed by projecting the data onto a two-dimensional plane, with the values of the series on the x-axis and their increments on the y-axis. Geometric properties of these phase portraits can be used to identify features and variations in the time series. This approach allows for a better understanding of short-term trends and fluctuations in economic and financial indicators.

AB - When analysing short-term economic and financial data, it is important to consider phenomena such as economic cycles, trends, and their changes, as well as structural shifts and financial bubbles. This study proposes an effective tool for visualizing these phenomena for high-frequency data, which refers to data collected on a weekly, daily, or more frequent basis. The method involves several steps. First, the choice of a time interval, the change within which is insignificant for analysis purposes. Next, the time series is smoothed using a filter that ignores these intervals. Then, the increments of the time series are calculated by taking differences between neighbouring values. Finally, a phase portrait is constructed by projecting the data onto a two-dimensional plane, with the values of the series on the x-axis and their increments on the y-axis. Geometric properties of these phase portraits can be used to identify features and variations in the time series. This approach allows for a better understanding of short-term trends and fluctuations in economic and financial indicators.

KW - Christina-Fitzgerald band-pass filter

KW - economic time series

KW - phase portrait

UR - https://www.mendeley.com/catalogue/3f78b590-452e-31aa-b0cb-2a1f9d0db570/

U2 - 10.1007/978-3-032-08603-7_27

DO - 10.1007/978-3-032-08603-7_27

M3 - Conference contribution

SN - 9783032086020

T3 - Lecture Notes in Business Information Processing

SP - 432

EP - 443

BT - Digital Economy. Emerging Technologies and Business Innovation. ICDEc 2025

PB - Springer Nature

Y2 - 15 May 2025 through 17 May 2025

ER -