Classifying Mutual Funds based on Relative Performance using Artificial Neural Networks
Abstract: This thesis investigates how well artificial neural networks perform when analyzing mutual funds to predict future performance. It does so from the viewpoint of a financial advisor, and uses a multilayer perceptron attempting to classify funds into three groups based on eight fund-specific variables. From both a technological and organizational perspective, the data sets and model used are incapable of predicting satisfactory results. Consequently, it is not applicable for a financial advisor in its current form. However, it is still plausible that neural networks could be used in investment analysis. Similar to other industries where technological advancements have had a disruptive eff ect, the finance industry could face a radical change resulting from a higher degree of robotic automation. Building on the insights of this study and conducting further research in the area should thus be considered important.
AT THIS PAGE YOU CAN DOWNLOAD THE WHOLE ESSAY. (follow the link to the next page)