ACADSTAFF UGM
| Title | : | Robust Mean–Variance Portfolio Selection Using Cluster Analysis: A Comparison between Kamila and Weighted K-Mean Clustering |
| Author | : |
LA GUBU (1) Prof. Dr.rer.nat. Dedi Rosadi, S.Si., M.Sc. (2) Dr. Abdurakhman (3) |
| Date | : | 19 2020 |
| Keyword | : | KAMILA clustering, Weighted k-means clustering, Robust estimation, FMCD estimation, S estimation, Outliers, Portfolio optimization. KAMILA clustering, Weighted k-means clustering, Robust estimation, FMCD estimation, S estimation, Outliers, Portfolio optimization. |
| Abstract | : | This studypresentsrobust portfolio selection usingclusteranalysisofmixed-type data. For this empiricalstudy, the daily price data of LQ45 index stocks listed on the Indonesia Stock Exchange were employed. First, six stocksclusters are formedby using the KAMILA algorithmon a combination of continuous and categorical variables. Forcomparisonpurposes, weighted k-means clusteranalysiswas also undertaken. Second, stocks that were representativeof each cluster, those with the highest Sharpe ratios,were selected to createa portfolio. The optimum portfolio was determined throughclassic (non-robust) and the robust estimation methodsoffast minimum covariance determinant (FMCD) and Sestimation. Using a robustprocedureenablesthe best-performingportfolio to be created efficiently when selecting assets froma large number of stocks, especially astheresultsarelargely unaffected inthe presence of outliers. This studyfoundthat the performance of the portfolio developed with theKAMILA clustering algorithm androbust FMCD estimationoutperformed those createdby other methods. |
| Group of Knowledge | : | Statistik |
| Original Language | : | English |
| Level | : | Internasional |
| Status | : |
Published
|