Module Directory

This module aims to provide a masters level module for students in Essex Business School which provides an up-to-date and rigorous course in financial time series econometrics.

This module will introduce students to the core methodology and empirical practice of modern financial time series econometrics.  This will involve developing relevant statistical and mathematical skills, together with the ability to apply these methods to real data using industry-standard software and to interpret their results.

The aims of this module are:

• To provide training in relevant statistical techniques required to perform robust research in financial econometrics, and to illustrate the applicability of these techniques to financial data.


• To provide students with hands on experience of implementing the methods commonly employed in the financial econometrics literature through practical sessions in which they will be trained in the use of relevant computing software.


• To develop and transmit knowledge of recent advancements in the financial econometrics literature through examining key papers that develop and/or utilise the econometric techniques covered in the course material.

By the end of this module, students will be expected to be able to:

1. Have an understanding of the relevant statistic techniques appropriate for performing both theoretical and empirical analysis in financial econometrics.


2. Have developed detailed knowledge on the use of a relevant statistical software package, being able to use this software to perform advanced statistical techniques on either user imported or pseudo-data.


3. Be able to critically evaluate contributions made by recent papers in the financial econometrics literature, and to replicate the findings outlined in these papers.

Syllabus

• Univariate Stationary Time Series I: Definition of time series processes; Stationarity and invertibility; Autocovariance and autocorrelation functions; Autoregressive Moving Average (ARMA) processes.


• Univariate Stationary Time Series II: The Wold decomposition; The lag operator; Common factors; Aggregation of time series processes; Estimation of ARMA models.


• Univariate Stationary Times Series III: ARMA Model selection - Identification, Diagnostic Checking, Information Criteria; Forecasting in ARMA models – Optimal Forecasts, Estimation Effects.


• Univariate Non-Stationary Time Series: Difference stationary (integrated) processes and trend stationary processes; ARIMA processes; Drifts and trends; Properties of unit root series; Spurious regressions.


• Unit Root and Stationarity Testing: (Augmented) Dickey Fuller unit root tests; Lag selection and deterministic components; KPSS stationarity tests.


• Rational bubble detection: Explosive autoregressive processes; Testing for bubbles; Historical detection and real-time detection of bubbles.


• Time Series Models of Volatility: Martingale difference processes; Conditional variance models: GARCH and ARSV models.


• Multivariate Time Series Models: VARMA, VMA and VAR models; VARIMA models; Granger causality.


• Cointegration and error correction models: Definitions; Interpretation; Estimation and Testing.


• Predictive Regression Modelling: Testing for predictability of financial returns; Dealing with persistence and endogeneity in the data.

This module will be delivered via:

• One 2-hour lecture per week.
• One 1-hour seminar per week.
• One 1-hour computer class per week.