/


Advanced ERM Session 2: Risk measures and Measuring, managing and mitigating market, credit and operational risk


This presentation is based on a part of an academic course on Advanced Enterprise Risk Management (Advanced ERM) titled ‘Risk measures and Measuring, managing and mitigating market, credit and operational risk’ and covers topics such as: mathematical underpins, market risk, credit risk, operational risk and managing market, credit and operational risk. It also includes appendices covering: VaR coherence and VaR vs TVaR, GARCH models, Maximum Likelihood Estimation (MLE), principal components analysis (PCA) and hazard rates and fitting operational risk loss distributions

Slides
1Session 2: Risk measures and Measuring, managing and mitigating market, credit and operational risk
2Session 2: Risk measures and Measuring, managing and mitigating market, credit and operational risk
3Model evolution characteristics
4Whatever we do requires
5Terminology
6Defining value and applicable axioms
7What if no suitable market in asset or liability?
8Worth noting implication on loss definition
9Assumed future behaviour of assets and liabilities
10More generally
11Most this course can realistically do
12Case study / scenario
13Selection of risk measures
14Advantages and disadvantages of VaR
15TVaR has certain technical advantages over VaR
16Mathematical definitions of VaR and TVaR
17VaR estimation
18VaR closely linked to capital requirements
19VaR and expected/unexpected loss
20Other risk measures, e.g.
21Marginal risk, e.g. marginal VaR (MVaR)
22The Gaussian case
23Incremental risk, e.g. incremental VaR (IVaR)
24Risk budgeting
25Statistical techniques for estimating risk measures
26Parametric approach
27Issues
28Incomplete or out-of-date data
29Non-linear exposures
30Call option price
31Heteroscedasticity
32Non-parametric approach: order statistics
33Standard errors in non-parametric approach
34Empirical studies of VaR estimates
35Back testing
36Basel back testing rules for banks
37Session 2: Risk measures and Measuring, managing and mitigating market, credit and operational risk
38Market risk
39Typical approaches (1)
40Typical approaches (2)
41New directions for market risk: IRC
42New directions for market risk: Stress VaR
43Session 2: Risk measures and Measuring, managing and mitigating market, credit and operational risk
44Credit risk
45Market developments
46Implications of new credit markets
47Credit portfolio risk models
48Ratings-based models
49Ratings-based models: components
50Latent variables
51Example
52Pros and cons of ratings based models
53Equity-based credit portfolio risk models
54Equity and debt as options on asset value
55Implementing an equity-based credit risk model
56Implementing the model statistically
57Default trigger
58Strengths and weaknesses of equity-based models
59Mixture models (1)
60Mixture models (2)
61Implementation issues for either type of model
62Broader issues
63Session 2: Risk measures and Measuring, managing and mitigating market, credit and operational risk
64Operational Risk
65Agency problems, monitoring and control
66More controls or more control?
67An illustrative game theoretic model of fraud
68Equilibria
69Characteristics of intermittent fraud equilibrium
70Managing operational risk in practice
71Benchmarking processes and losses versus others
72Sharing data
73Capital for Operational Risk
74Insurance as a substitute for capital
75Session 2: Risk measures and Measuring, managing and mitigating market, credit and operational risk
76Managing market, credit and operational risk
77Any action to reduce one risk can increase another
78Quantifying risk transformation risks
79Main 'market' hedging instruments
80Dynamic hedging (1)
81Dynamic hedging (2)
82Risk/reward analysis and asset/exposure allocation
83Session 2: Agenda covered
84Session 2: Risk measures and Measuring, managing and mitigating market, credit and operational risk
85Appendix A: VaR coherence and VaR vs TVaR
86Axiomatic approach: coherence
87Characterisation of coherent risk measures
88VaR is coherent for Gaussian distributions
89Arguments favouring TVaR based on coherence
90More generic arguments favouring TVaR
91Which takes into account loss in the event of default?
92Different stakeholder perspectives (1)
93Different stakeholder perspectives (2)
94Treatment of illiquidity (1)
95Treatment of illiquidity (2)
96Treatment of illiquidity (3)
97Stress testing methodologies
98Appendix B: GARCH models
99Properties of GARCH models
100RiskMetrics approach
101RiskMetrics specification
102Appendix C: Maximum Likelihood Estimation
103Log likelihood
104Desirable properties of ML estimator
105Likelihood ratio tests
106Binomial back testing
107Introducing a restriction
108Appendix D: Principal components analysis
109Principal components - explanation or noise?
110Smaller eigenvalues/principal components
111Appendix E: Hazard rates and fitting loss distributions
112Hazard rates
113Simple example of a hazard rate
114Hazard rates that vary through time
115Simulating independent failure times
116Incorporating dependence in failure times
117Fitting loss distributions
118E.g. Gamma distribution
119Important Information



NAVIGATION LINKS
Contents | Next | ERM Lecture Series


Desktop view | Switch to Mobile