PENGUKURAN VALUE-AT-RISK DENGAN VOLATILITAS TAK KONSTAN DAN EFEK LONG MEMORY

• Main Authors: , Sukono, Drs.,M.Si, , Prof. Drs. Subanar, Ph. D.
• Format: Theses and Dissertations NonPeerReviewed
• Published: [Yogyakarta] : Universitas Gadjah Mada 2011
• Subjects: ETD
• Online Access: https://repository.ugm.ac.id/90577/; http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=53075
• Institution: Universitas Gadjah Mada

Variance or standard deviation as a measure of risk can not always accommodate all risk of loss events occur. Emerged the idea of risk measurement is done by using quantile, which gave birth to the risk measure of Value-at-Risk. It is estimated that the risk measurement Value-at-Risk will continue to grow rapidly, especially those using the approach of non-constant volatility and there is a long memory effect. In this dissertation is a research on the measurement of investment risk using the basic framework of the Value-at-Risk (VaR). It is assumed that stock returns have a mean and non constant volatility, and there is long-memory effect. Other assumptions that stock returns follow the pattern of Capital Asset Pricing Model (CAPM). Theoretically, it has been developed the theorem of Modified Value-at-Risk (MVAR) in Student-t distribution. Under the CAPM pattern, it has been developed the VaR measurement theorem of under CAPM with lagged, and the theorem of VaR under CAPM of Koyck transformation. Using the basic framework of measurement risk of VaR, has also been developed the theorem of weight in the mean-VaR portfolio optimization. In this dissertation, the identification of long memory effect conducted using the method of rescale range (R/S) or Geweke and Porter-Hudak (GPH). The mean and non constant volatility is estimated using the ARFIMA-GARCH models. To measure the performance of VaR performed using statistical of quadratic probability score (QPS) by Lopez model II. The purpose of this research is to develop an alternative method in accordance with the characteristics of stock returns data. The application of the development of the theoretical results in empirical research, used to analysis the data of 10 (ten) shares, Composite Stock Price Index (IHSG), and riskfree asset (bond).