Realized GARCH and Real GAS, Realized Kernel from the article Barndorff-Nielsen, Hansen, Lunde, Shephard. (2011) you can get info on that. At the parameter-driven part (SV models in time series) I want information (5) pages on the Local level model, Quasi Maximum Likelihood, Indirect Inference, Importance Samling, and Particle Filtering. From e.g. book: (Time Series Analysis by State Space Methods by Durbin and Koopman) And I want some information (2 pages) about the Gaussian distribution and Student-t distribution in this setting. Because the modeling will be done in these two distributions. Chapter Data cleaning (2 pages). Information on how to clean high-frequency trading data (Barndorff-Nielsen et al.(2009)). Chapter Methodology (~8 pages): How to implement the estimation methods for modeling/forecasting volatility with steps and formulas. For observation driven models: Maximum Likelihood estimation (in both distributions) For parameter-driven models: Kalman Filter, Kalman Filter Smoother, Quasi Maximum Likelihood, Indirect Inference (Eric Renault et al 1993), and Importance sampling.
