Consumer Purchase Dynamics of 190,000 Individuals Before, During, and After COVID-19

Yang Tian from Tsinghua University provides a dataset for analyzing scaling and criticality in large-scale social systems. It contains approximately 2.2 million purchase orders made by around 190,000 individuals from January 2018 to December 2022. The data describes the daily consumption frequencies for about 6,000 commodity types.

Use Cases

• Analyzing the impact of the COVID-19 pandemic on consumer behavior based on purchase order data.
• Studying scaling laws and critical transitions in social systems using decomposed trend, seasonal, and random consumption components.
• Modeling long-term and short-term consumption dynamics across approximately 6,000 commodity types.
• Applying renormalization group methods to time-series data of large-scale social systems.

Strengths

• Contains approximately 2.2 million purchase orders, providing a substantial base for analysis.
• Covers a 5-year time range from January 2018 to December 2022, capturing pre-, during, and post-pandemic periods.
• Tracks consumption dynamics for about 6,000 distinct commodity types.
• Includes data from around 190,000 individuals, offering a large-scale view of social behavior.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Commodity names are written in Chinese, which may require translation for non-Chinese speakers.
• Row count for the derived time-series file is unknown, which may limit suitability assessment.

Provenance

Source

Tsinghua University

Collection Method

Naturalistic observation of daily consumer behaviors derived from purchase orders.

Time Range

January 2018 to December 2022