Assimilation of PSO and SVR into an improved ARIMA model for monthly precipitation forecasting
2024

Improved ARIMA Model for Monthly Precipitation Forecasting

Sample size: 4 publication 10 minutes Evidence: high

Author Information

Author(s): Parviz Laleh, Ghorbanpour Mansour

Primary Institution: Faculty of Agriculture, Azarbaijan Shahid Madani University, Tabriz, Iran

Hypothesis

Can an improved ARIMA model using SVR and PSO provide more accurate monthly precipitation forecasts compared to traditional methods?

Conclusion

The improved ARIMA model significantly enhances the accuracy of monthly precipitation forecasts, outperforming conventional models.

Supporting Evidence

  • The improved ARIMA model showed a 39.92% reduction in normalized mean squared error compared to the original ARIMA.
  • Using SVR in the modeling process improved forecast accuracy at all stations.
  • The residual predictive deviation of the improved model was greater than 2, indicating high performance.

Takeaway

This study created a better way to predict monthly rainfall using a special math model that combines different techniques to make it more accurate.

Methodology

The study used an improved ARIMA model that incorporates support vector regression (SVR) and particle swarm optimization (PSO) for forecasting monthly precipitation at four stations in northern Iran.

Limitations

The study is limited to four specific stations in northern Iran, which may not represent other regions.

Participant Demographics

The study focused on precipitation data from four weather stations in northern Iran.

Digital Object Identifier (DOI)

10.1038/s41598-024-63046-3

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication