Welcome to the Bahria University DSpace digital repository. DSpace is a digital service that collects, preserves, and distributes digital material. Repositories are important tools for preserving an organization's legacy; they facilitate digital preservation and scholarly communication.
dc.contributor.author | Kumail Abbas, 01-397201-009 | |
dc.date.accessioned | 2022-01-10T10:55:08Z | |
dc.date.available | 2022-01-10T10:55:08Z | |
dc.date.issued | 2021 | |
dc.identifier.uri | http://hdl.handle.net/123456789/11545 | |
dc.description | Supervised by Dr. Muhammad Anees Khan | en_US |
dc.description.abstract | vii ABSTRACT Macroeconomic indicators are the most important indicators of an economy. The application of new machine learning methods has led to introducing new and improved intelligent methods for solving time series forecasting problems in various scientific fields. This study interested in investigated the performance ML algorithms. For that, compared different machine learning methods with each other, which are ANN, KNN, SVM and Polynomial regression in forecasting inflation rate and exchange rate (PKR/ USD) for Pakistan. The sample period is from Jan 1989 to Dec 2020. The data set is split into two sets the data from Jan 1989 to Dec 2018 for the training of machine algorithms and the remaining data for model testing. For accuracy of the forecast are measured by using RMSE and MAE. During forecasting of inflation rate based on prediction error the test set shows that the polynomial regression degree 1 and ANN outperform the SVM and KNN. During forecasting of exchange rate, SVM rbf outperforms the KNN, Polynomial and ANN. | en_US |
dc.language.iso | en | en_US |
dc.publisher | Management Studies BUIC | en_US |
dc.relation.ispartofseries | MS (Fin);MFN-T 9507 | |
dc.subject | MS Finance | en_US |
dc.subject | Machine Learning | en_US |
dc.subject | Prediction Error ANN, SVM, KNN MSE, RMSE | en_US |
dc.title | Artificial Intelligence In Relation To Macroeconomic Factors For Predicting The Future | en_US |
dc.type | MS Thesis | en_US |