Enhancing Imputation Performance in Univariate Time Series Using Gated Recurrent Unit

Missing data is a pervasive problem in time series analysis, often leading to biased predictions and compromised model performance. This study introduces the Gated Recurrent Unit (GRU) as a robust solution for estimating missing values in time series data. Experimental results demonstrate GRU's superior performance compared to traditional machine learning methods like Random Forest, Support Vector Regression, Extra Trees, and K-Nearest Neighbors. GRU's exceptional ability to capture complex, dynamic relationships in time series data makes it an ideal choice for accurate imputation, ensuring the reliability of predictive models for time series data.