RDP 1999-04: Value at Risk: On the Stability and Forecasting of the Variance-Covariance Matrix Equation (7)

\begin{array}{l} σ_{i, t + 1}^{2} = ω_{i} + α_{i} r_{i, t}^{2} + β_{i} σ_{i, t}^{2} \\ = \frac{ω_{i}}{1 - β_{i}} + α_{i} \sum_{j = 0}^{T} β_{i}^{j} r_{i, t - j}^{2} \end{array}