Crop-Yield-Prediction

Crop yield prediction is a crucial factor in ensuring the profitability of farmers. Predicting future crop yields can help farmers make informed decisions about planting and harvesting and optimize their crop yield. In this project, we used R to analyze historical data on crop yields and weather patterns to predict future yields. To best advise farmers, we focused on predicting the crop yields of the ten most consumed crops in the world. Doing so will make our predictions applicable to a larger population asthese are produced in great quantities to fulfill worldwide demand.

Four models were used to make predictions and their RMSE values were calculated:

• Linear regression (with an RMSE value of 46,238.99)
• Regression tree (with an RMSE value of 84,181.4009)
• XGBM (with an RMSE of 47008.54)
• GBM (with an RMSE of 27325.66)

The results suggest that GBM had the lowest RMSE, followed by XGBM, linear regression, and regression tree, indicating that the gradient boosting method was the most accurate model for making predictions.