Abstract:
The rate of return is one of the important factors for investors in making decisions when
they want to invest. The amount of return obtained can determine the level of price
fluctuations over a certain period or can be called volatility. Stock volatility needs to be
reviewed to measure how big the level of fluctuation in the price of a stock is, so in this
study discussed the analysis of the volatility of PT. AMAG uses the Generalized
Autoregressive Conditional Heteroscedasticity (GARCH) method. The results showed
that the stock return log data of PT. AMAG has volatility as evidenced by the presence
of elements of heteroscedasticity in the Autoregressive Integrated Moving Average
(ARIMA) model, so it is necessary to continue with the GARCH model testing to
overcome the problem of heteroscedasticity or the residual variance is not constant. The
best models obtained are ARIMA (0,0,1) and GARCH (1,1), using these models the
results of forecasting the stock price index of PT. AMAG for 30 periods did not differ
much from the actual data with a MAPE value of 1.44%.
