Modelling irregulary spaced financial data. Theory and practice of dynamic duration models. (Q1880662): Difference between revisions

This book regards financial point processes. A salient feature of financial event data is the occurent of strong clustering structures. The modelling of such data requires dynamic point process approaches. By selecting the points of a process in a certain way, specific types of financial point processes are generated. So, the analysis of trade duration allows a deeper insight into market dynamics. Valuable risk and liqudity measures are constructed by defining financial events in terms of price and/or the volume process. Several applications are illustrated. This book has 8 chapters. After the introduction (chapter 1), chapter 2 and 3 is devoted to the provision of the methodological and economic background. Chapter 4 describes the databases. The chapters 5--7 present the different econometric models and illustrate their application to various financial data. A summary and conclusions follow in chapter 8. After chapter 1 (Introduction), chapter 2 (Point Processes) regards basic concepts of, types of, and nondynamic, point processes. Censoring and time-varying covariates and an outlook on dynamic extensions follow. Chapter 3 (Economic Implications of Financial Durations) describes types of financial durations, the role of trade durations in market microstructure theory, risk estimation based on price durations, and liquidity measurement. Chapter 4 (Statistical Properties of Financial Durations) treats data preparing issues, transaction databases and data preparation. Statistical properties of trade, limit order, quote duration follow. Then statistical properties of price durations and volume durations are treated. It ends with a summary of the statistical findings. Chapter 5 (Autoregressive Conditional Duration Models) begins with the ARMA models for (log-) durations. Then the autoregressive conditional duration (ACD) model is described. Some extensions, tests, and applications of this model follow. After the dynamic integrated intensity processes, chapter 6 (Semiparametric Dynamic Proportional Intensity Models) regards the semiparametric autoregressive conditional intensity (ACPI) model. After describing the model, properties, extensions, and tests of the ACPI model are treated. The chapter ends with estimating volatility using The ACPI model. Chapter 7 (Univariate and Multivariate Dynamic Intensitiy Models) begins with the description of the univariate and multivariate dynamic intensity model. Dynamic latent factor models for intensity processes and applications of dynamic intensity models follow. Chapter 8 gives a summary and conclusions.