VALUE at RISK (VaR) DAN CONDITIONAL VALUE at RISK (CVaR) DALAM PEMBENTUKAN PORTOFOLIO BIVARIAT MENGGUNAKAN COPULA GUMBEL

Dina Rahma Prihatiningsih (Departemen Statistika, Fakultas Sains dan Matematika, Universitas Diponegoro)
Di Asih I Maruddani (Departemen Statistika, Fakultas Sains dan Matematika, Universitas Diponegoro)
Rita Rahmawati (Departemen Statistika, Fakultas Sains dan Matematika, Universitas Diponegoro)

Jurnal Gaussian
Vol 9, No 3 (2020): Jurnal Gaussian

One way to minimize risk in investing is to form of portfolio by combining several stocks.Value at Risk (VaR) is a method for estimating risk but has a weakness that is VaR is incoherent because it does not have the subadditivity. To overcome the weakness of VaR, Conditional Value at Risk (CVaR) can use. Stock data is generally volatile, so ARIMA-GARCH is used to model it. The selection of ARIMA models on R software can be automatically using the auto.arima() function. Then Copula Gumbel is a method for modeling joint distribution and flexible because it does not require the assumption of normality and has the best sensitivity to high risk so that it is suitable for use in stock data.The first step in this research is to modeling Copula Gumbel-GARCH with the aim to calculate VaR and CVaR on the portfolio of PT Bank Mandiri Tbk (BMRI) and PT Indo Tambangraya Megah Tbk (ITMG). At the confidence level 99%, 95%, and 90% obtained the VaR results sequentially amounted to 3.977073%; 2.546167%; and 1.837288% and the CVaR results sequentially amounted to 4.761437%; 3.457014%; and 2.779182%. The worst condition is a loss with VaR and it is still possible if a worse condition occurs is a loss with CVaR so that investors can be more aware of the biggest loss that will be suffered.Keywords: Value at Risk, Conditional Value at Risk, Auto ARIMA, Copula Gumbel.