Enhancing mortgage rate prediction: a comprehensive evaluation of computational statistical approaches

Abstract

The noterate is a tool for predicting home mortgage rates, it is often skewed and has missing information. The noterate could be affected by incomplete or inaccurate data, thus leading to inaccurate predictions. Financial organizations including mortgage companies or banks need to consider the risk of uncertainty carefully and make a more accurate prediction based on some suitable models. To deal with this situation, in this paper we compared six computational statistical methods, including the ordinary least square model, maximum likelihood estimation, maximum a posterior, bootstrapping, Metropolis-Hastings, and Gibbs sampling method on a mortgage dataset. Based on the k fold cross-validation technique and four metrics including mean absolute error (MAE), mean squared error (MSE), root mean squared error (RMSE), and mean absolute percentage error (MAPE), the bootstrapping method outperforms other methods. In practice, this method is recommended for predicting noterate.

Keywords:

• Bootstrapping
• maximum likelihood estimation
• maximum a posterior
• Gibbs sampling
• Metropolis-Hastings

AMS Subject classifications:

• 91-10
• 91-11
• 68T05
• 68T09
• 68T10

Acknowledgments

We would like to express our sincere gratitude to the reviewers for their thoughtful comments and valuable suggestions that greatly contributed to improving the quality of this paper. We also extend our thanks to the editors for their guidance and support throughout the publication process.

Disclosure statement

The authors declare that they have no conflict of interest. All authors have contributed equally to this work. All authors agreed to the submission and potential publication.