Time Series Analytics: Estimation, Forecasting, and Volatility — Volume II (Time Series Analytics: Theory and Python Practice Book 2)

The second volume of a complete graduate course on time series, building directly on Volume I. Volume II covers, in twenty-two chapters: conditional-expectation forecasting; the innovations and Durbin–Levinson algorithms; maximum-likelihood and conditional sum-of-squares estimation; AIC, BIC, and Hannan–Quinn model selection; residual diagnostics; the Dickey–Fuller, Phillips–Perron, and KPSS unit-root tests; vector autoregression with Granger causality, impulse-response functions, and FEVD; Engle–Granger cointegration; and the ARCH, GARCH, EGARCH, and GJR-GARCH families with backtested Value-at-Risk on real Indian-equity, gold, and BTC-USD data. Each chapter includes two theorems with full proofs, two worked examples (one synthetic-and-hand-computable, one drawn from real markets), runnable Python with seeded numerical output, and five graded discussion questions with full solutions. Read more