The prediction of import and export of commodities has occurred between countries to buy or sell goods essential for humans. Governments need to keep track of the number of imports or exports to ensure the increase of their country's Gross Domestic Product (GDP). Support Vector Machine (SVM) is a robust classification algorithm to classify data efficiently. Support Vector Regression (SVR) is a modification of SVM that predicts absolute values. Fine-tuning the parameters of SVR is not easy, and it is hard to visualize their impact on the dataset. SVR uses the support vectors obtained during the running of the algorithm to predict the dataset's outcome. The new version of the SVR algorithm is proposed, assisted with modified RBF Kernel to improve the model's efficiency. The paper's main contribution is towards the field of economic data analysis, as in, to predict the commodities of goods imported and exported for each country. The purpose of this paper is to use SVR in a commodity dataset to predict each commodity's price being imported and exported for limited countries and encourage the use of machine learning in economics. Further, LSTM is applied for prediction in layers to predict the weight of some incoming commodities to countries. We then obtain the expected results and find the model's accuracy using this result over a real dataset. The problems faced during the research were that data were not available for some countries and information for commodities and countries was not uniform throughout the years. Due to this, the model did not fit the data accurately for those countries. The interpretation of the model shows that the overall error for the proposed model is very trivial and hence produces higher accuracy. To conclude, the predictions were accurate using SVR.