OPTIMISASI PORTOFOLIO ROBUST MENGGUNAKAN SECOND-ORDER CONE PROGRAMMING (SOCP)

• Main Authors: , DESSY PARAMITA, , Dr.rer.nat. Dedi Rosadi, M.Sc.
• Format: Theses and Dissertations NonPeerReviewed
• Published: [Yogyakarta] : Universitas Gadjah Mada 2013
• Subjects: ETD
• Online Access: https://repository.ugm.ac.id/126001/; http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=66186

Portfolio optimization is one of the best known and most widely used methods in financial portfolio selection. The first portfolio optimization technique called mean-variance model was developed by Harry Markowitz (1952). Despite the strong theoretical support and the availability of efficient computation provided by mean-variance, the model presents several practical pitfalls. One of them is that the model is often sensitive to the change in input parameter. To reduce the sensitivity of mean-variance model, the robust portfolio optimization technique has been proposed. In this approach, the input parameter are expected to lie within a confidence interval, which is described as uncertainty sets. This is because in reality, it is very difficult to estimate the correct values of these parameters and the values change every time. After determining the uncertainty sets, the analysis is carried out for the worst-case scenario under the model, i.e: model with minimum expected return and maximum risk. The optimization problem is reduced to a second-order cone programming (SOCP) which could be solved via primal-dual interior point method. The case study presents the portfolio construction of several stocks listed in Indonesia Stock Exchange i.e: AALI, ADRO, ASRI, BBRI, CPIN, TLKM and UNVR, using the SOCP and mean-variance optimization techniques. These two methods are compared by measuring the portfolio performance under their rate of return and Sharpe ratio. As aresult, the SOCP robust portfolio performs better than the mean-variance portfolio.