Historical risk data can be used to apply predictive modeling to forecast future states of risk, as well as Risk Tracking, Time-Series Risk Forecasts, PDF/CDF Likelihood of Occurrence, and Snapshots per period and over time (Figure 2.15). In the Risk Forecast section, using either historical data or subject matter estimates, you can run forecast models on time-series or cross-sectional data by applying advanced forecast analytics such as ARIMA, Auto ARIMA, Auto Econometrics, Basic Econometrics, Cubic Splines, Fuzzy Logic, GARCH (8 variations), Exponential J-Curves, Logistic S-Curves, Markov Chains, Generalized Linear Models (Logit, Probit, and Tobit), Multivariate Regressions (Linear and Nonlinear), Neural Network, Stochastic Processes (Brownian Motion, Mean-Reversion, Jump-Diffusion), Time-Series Analysis, and Trendlines.

error: Content is protected !!