Asian Journal of Probability and Statistics

Weather and climate have a significant impact on our daily lives. The amount of data available is vast and in order to draw meaningful conclusions from the data one needs to develop appropriate models that can fit the data properly. Several models are available for analyzing and predicting the time series data. The spline-based technique produces an acceptable pattern of time series data that can be used for prediction and modelling. The data for the study is obtained from the Government of India website for the meteorological subdivision of East and West Uttar Pradesh for the past 120 years from 1901 to 2020 on a monthly basis. This study discusses the application of Autoregressive Integrated Moving Average (ARIMA) model and Natural Cubic Spline for the development of a more appropriate model for predicting and forecasting rainfall in Uttar Pradesh. The study focuses on two different approaches of applying the cubic spline. The fitted models are compared by plotting the curves and various matrices like Root Mean Square Error (RMSE), R square, adjusted R square, AIC and BIC. The analysis was performed using R software. From the comparison, the model fitted through Natural Cubic Spline is obtained as more suitable for the prediction and forecasting of Rainfall in Uttar Pradesh. With a better model for forecasting of rainfall, the study will contribute to better decision-making for policymakers, thus eventually contributing to better crop management in the state. The study will also contribute to better water resource management and enhance disaster preparedness. The study can also serve as a foundation for future research on rainfall prediction in UP and other regions and encourage further exploration of different statistical techniques and data sources.