#TITLE_ALTERNATIVE#

• 主要作者: SHABRINA ( NIM : 10109032 ) Pembimbing : Khreshna I.A. Syuhada, M.Sc, Ph.D, AYU
• 格式: Final Project
• 語言: Indonesia
• 在線閱讀: https://digilib.itb.ac.id/gdl/view/18124

Extreme value theory is a cornerstone of a method to identify extreme values <br /> <br /> <br /> <br /> <br /> and model it in a distribution. In the financial field, extreme values can occur <br /> <br /> <br /> <br /> <br /> in the decline of stock price index. Extreme values are rare but certainly have a <br /> <br /> <br /> <br /> <br /> great impact the risk of extreme losses. Efiorts to control risk is to understand <br /> <br /> <br /> <br /> <br /> and quantify the risks appropriately. Value-at-Risk (VaR) is a popular measure of <br /> <br /> <br /> <br /> <br /> risk used in financial risk management. Thus, in determining the risk of extreme <br /> <br /> <br /> <br /> <br /> value are used necessary calculations using VaR and Extreme Value Theory. Ex- <br /> <br /> <br /> <br /> <br /> treme Value Theory with Block Maxima Method identify extreme values as the <br /> <br /> <br /> <br /> <br /> maximum value of data per block period of time, thus determining the block size <br /> <br /> <br /> <br /> <br /> in this method is very important. The extreme values are modeled by a family <br /> <br /> <br /> <br /> <br /> of GEV distribution. Extreme value of the stock index's decline can be modeled <br /> <br /> <br /> <br /> <br /> with distribution Frechet. Parameter estimator of the extreme value distribution <br /> <br /> <br /> <br /> <br /> can be used in the calculation of VaR. From the calculation, the value of VaR <br /> <br /> <br /> <br /> <br /> is in <br /> <br /> <br /> <br /> <br /> uenced by the block size and the chosen level of confidence. So extreme <br /> <br /> <br /> <br /> <br /> decline in the stock index will be greater as the larger block size and confidence <br /> <br /> <br /> <br /> <br /> level.