Economic analysis using higher-frequency time series: challenges for seasonal adjustment

Area of Science:

• Economics
• Econometrics
• Data Science

Background:

• The COVID-19 pandemic heightened the demand for real-time economic indicators.
• Higher-frequency data (daily, weekly) are crucial for timely economic assessment.
• Seasonal and calendar adjustment of these time series presents significant challenges.

Purpose of the Study:

• To analyze the specific features and challenges of seasonal adjustment for high-frequency economic time series.
• To develop a classification (taxonomy) of these features.
• To discuss potential adjustment options and difficulties.

Main Methods:

• Analysis of German high-frequency time series: hourly electricity consumption, daily truck mileage, and weekly Google Trends data.
• Examination of time series characteristics relevant to seasonal adjustment.
• Development of a taxonomy for seasonal high-frequency time series.

Main Results:

• Identified unique features and idiosyncrasies of high-frequency economic time series impacting seasonal adjustment.
• Provided a structured taxonomy categorizing these features.
• Discussed practical obstacles and potential solutions for adjustment.

Conclusions:

• Effective seasonal adjustment of high-frequency economic data requires understanding specific time series characteristics.
• The developed taxonomy aids in addressing the complexities of real-time economic monitoring.
• Accurate adjustment is vital for reliable economic assessment during crises.