Calculation of Value at Risk (VaR) based on Autoregressive Conditional Heteroscedasticity (ARCH) model.

Abstract

Time commercial enterprises appear and provide more effective tools and devices for the future of the enterprises to solve dilemma of profit and create new profit growth. Value at Risk (VaR) model is used to solve the problems of time commercial enterprises. VaR model is a model used to measure the trade risk such as market risk, credit risk and so forth. The risk of VaR model refers to certain level of trust, in the period of the biggest loss probability. Data modelling of time series is done by variant assumption of constant error (homoscedasticity). In fact, many time series data have error variant that is inconstant (heteroscedasticity), especially for time series data in finances. It causes the modelling with the assumption of homoscedasticity is not used. Engle (1982) develops a model whose lines and kinds of time series data are modelled simultaneously.  The model is know as Autoregressive Conditional Heteroscedasticity (ARCH) model. ARCH model is applied in share close price of Bank Mandiri to determine Value at Risk (VaR).