Missing data is a big problem in data analysis and machine learning, and it can seriously affect the precision and efficacy of prediction models. The novel hybrid method of imputation is presented to address this issue and hopefully overcome the difficulty. The approach predicts and fills in missing values in datasets by combining the advantages of four imputation techniques, such as statistical and machine learning-based imputation. By leveraging the synergistic power of these approaches, the hybrid imputation method offers a robust and versatile solution for handling missing data, ensuring the reliability of downstream analysis, and enhancing the overall quality of the predictive model. This imputation strategy is compared with four imputation strategies, specifically Mean, Mode, Multivariate Imputation by Chained Equations (MICE), and K-nearest neighbor (KNN) strategy. The performance of this method is evaluated on the Cleveland heart disease dataset using a Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF). The Hybrid Method of Imputation (HMI) has provided superior results than other existing strategies. The proposed approach has made a difference in expanding precision and decreasing the RMSE of classifiers.