Course 2025-2026 a.y.

1. Introduction and review of key concepts: Loss functions and decision theory; forecast evaluation

2. Forecasting stock returns; time-varying parameter models; hints to Bayesian Methods

3. Volatility modeling and forecasting

4. The instability of correlations: Multivariate GARCH and DCC models

5. The instability of correlations: Models With Breaks, Recurrent Regime Switching, and Nonlinearities

6. The instability of correlations: Markov Switching models

7. Realized volatility and covariance modelling

8. Climate risk in asset management

9. Bio-diversity risk in asset management

10. The role of structured products in dynamic asset management (4 hours)

• Explain decision‐theory principles and forecast evaluation metrics, including loss functions (e.g., MSE, MAE) and criteria for assessing predictive accuracy in financial contexts .

• Describe time‐varying parameter models for stock return forecasting, covering Bayesian estimation and key empirical findings from predictive‐regression literature .

• Explain univariate and multivariate volatility frameworks, such as ARCH/GARCH, DCC, and other multivariate GARCH specifications, to capture time‐varying risk and correlation dynamics .

• Analyze regime‐switching and nonlinear correlation models, including Markov‐switching and structural‐break approaches, to address unstable dependencies across assets .

• Understand realized volatility measures and their applications, and integrate climate‐ and biodiversity‐risk factors into portfolio construction and asset pricing decisions .

• Explain the role of structured products and derivative strategies in dynamic asset management, including how volatility‐based instruments (e.g., VIX futures) and structured notes can be used to enhance portfolio performance and tailor risk exposures .

• Estimate and evaluate forecasting models using decision‐theory principles, selecting appropriate loss functions and forecast‐evaluation metrics to judge predictive accuracy and guide investment decisions .

• Fit and interpret univariate and multivariate volatility models, such as ARCH/GARCH and DCC specifications, to measure time‐varying risk and produce out‐of‐sample volatility forecasts for portfolio risk management .

• Detect and model regime shifts and nonlinear dependencies in asset returns, applying Markov‐switching, structural‐break, and regime‐switching approaches to capture correlation instability in multi‐asset portfolios .

• Integrate climate and biodiversity risk factors into dynamic asset allocation, using ESG‐related measures and constructing volatility‐based or structured‐product strategies (e.g., VIX futures, volatility‐linked notes) to enhance portfolio performance under environmental risk considerations .

• Lectures
• Guest speaker's talks (in class or in distance)
• Individual works / Assignments

Lectures: Instructor‐led sessions (42-44 h) covering decision theory, volatility frameworks (ARCH/GARCH, DCC), regime‐switching models, realized‐volatility measures, climate/ESG risk, and structured‐product strategies. Lecture slides (Blackboard) follow Guidolin & Pedio, with board‐work examples that demonstrate both formal derivations and real‐world applications.

Guest Speaker’s Talks (1–2 sessions)
Invited practitioners present on climate‐risk integration and volatility‐based asset management (e.g., VIX products), linking lecture concepts to live market data or code demonstrations. Pre‐reads are provided so students arrive prepared to discuss how theoretical models serve professional needs.

Individual Works / Assignments
Students choose either a 12‐page literature review (up to 35 % of grade) on an assigned topic (e.g., ESG premia) or an empirical replication/extension project (up to 50 % of grade) approved by the instructor.

• Written individual exam (80-minute, open‐book/open‐notes; 100 % default weight): Consists of 3–4 essay‐type questions and numerical exercises that require students to explain decision‐theory concepts, interpret ARCH/GARCH output, assess unit‐root tests, and perform simple model calculations. This format verifies theoretical mastery (via essays) and practical skills (via numerical tasks). To pass, students must correctly address both conceptual explanations and applied computations.
   

• Individual Works/Assignments (optional, 35 %–50 % weight):

  • Literature review (12 pages; 35 %) tests ability to synthesize key papers on topics (e.g., ESG premia), demonstrating depth of understanding.

  • Empirical project (up to 50 %) requires replication/extension of a syllabus paper: data processing, model estimation (e.g., DCC‐GARCH), and written critique. Grading is based on correctness of implementation, rigor of analysis, and clarity of presentation.

  • A combined maximum of 80 % weight applies if both options are chosen.

The material covered in the course is outlined in lecture slides made available via the class Blackboard. Lecture notes and class presentations of the material should be taken as a guidance for further study on selected parts of the textbooks:

Guidolin, M., and M., Pedio (2016), Essentials of Applied Portfolio Management, EGEA and Bocconi University Press (EAPM).

Guidolin, M. and M., Pedio (2018) Essentials of Time Series for Financial Applications, Academic Press (ETSFA).

The following textbook may also be of some use:

Campbell, J. Y. (2017). Financial Decisions and Markets. Princeton University Press.

For each topic we will also provide suggestions for further reading, whose consultation is left to the students’ initiative.