Prediction of Industrial Property Rights Applications using Vector Autoregressive Model

Abstract

This study predicts the volume of industrial property applications using a vector autoregressive (VAR) model that incorporates macroeconomic factors such as economic growth, R&D, inflation, and interest rates. By experimenting with multiple variable combinations, 24 VAR models were derived, and the optimal model with the smallest prediction error was selected based on the forecast accuracy measured by the Mean Absolute Percentage Error (MAPE). Using this optimal VAR model, we forecast industrial property applications for the next ten years (2023-2032). The results indicate that industrial property applications are projected to increase annually by 5.01%, with patents increasing by 3.61%, utility models decreasing by 3.29%, designs increasing by 0.72%, and trademarks increasing by 1.96%.

Keywords

• vector autoregressive model
• time series analysis
• patent
• design
• trademark
• prediction