| Title |
Investicinio portfelio formavimo strategijos sukūrimas ir tyrimas / |
| Translation of Title |
Development and research of investment portfolio selection strategy. |
| Authors |
Jonaitis, Robertas |
| Full Text |
|
| Pages |
43 |
| Keywords [eng] |
investment portfolio ; stock market prediction ; ensemble methods |
| Abstract [eng] |
Profitable investment portfolio development requires a lot of knowledge and market analysis. Machine learning algorithms and optimization methods can facilitate this process. The paper describes financial ratios, portfolio development and evaluation tools. Approaches to asset selection are discussed in greater detail. Random Forest and Gradient Boosting regressors with the fundamental data are used to predict the changes of stock prices for the next quarter. Based on selected stocks, investment portfolios are developed utilizing various strategies. Portfolio weights are calculated using historical efficient frontier. Different strategies are tested in a time horizon of two years, revising portfolio in every quarter. The performance of portfolios is determined by the chosen benchmark and different quality metrics. |
| Dissertation Institution |
Kauno technologijos universitetas. |
| Type |
Master thesis |
| Language |
Lithuanian |
| Publication date |
2019 |
