FRM & MATLAB: Review of Systematic & Rule-Based Risk Management

Add to Favourites
Post to:

FRM & MATLAB: Review of Systematic & Rule-Based Risk Management

Type: ppt

Presentation Transcript Presentation Transcript

FRM & MATLAB Review of Systematic & Rule-Based Risk Management : FRM & MATLAB Review of Systematic & Rule-Based Risk Management Version 1 (for internal Use Only) How to apply Financial Risk management exam Studies and matlab Presentation and video Made for research & education By Satyadhar joshi 1

Content : Content Readings Introduction Different Type of Risks/Returns Hedging Efficiency Overlay Management Cost / Valuation / Risks Deciding Risk Techniques Interest Rate Models Interest Rate Payer Swap / Swap Options CDX Index Options Variance Swaps Options on VIX Implementation Dynamic Hedging Runtime Analysis Optimization Problem 2

Content (Contd…) : Content ( Contd …) Portfolio Risk Liquidity Risk Currency Options Equity Options Carbon Risk Inflation Risk BBB risk management Copula / Correlation Strategic allocation / Risk adjusted returns / VAR Methods Distributions MATLAB Toolboxes Used SQL in MATLAB Bloomberg API C# for Financial Engineering Summary References 3

Audience & Readings : Audience & Readings Developed for FRM Course takers & MATLAB Course takers MATLAB for Financial Engineering – A course by Qcfinance Indore FRM Level 2 Online classes Use in Pension or other long term Investment These assets can be given to Hedge funds on rent to play games in Algo trading 4

Introduction : Introduction Review of State-of-Art Research on MATLAB for Risk Sensitization of how to approach the area How things are linked up / Passive strategy / Index Quant Managing risk involved many complex derivatives Pricing and predicting our portfolio and managing risk budget remains a challenge Assumptions used is interest rate structure, geometric Brownian motion, normal distribution of equity prices 5

Risk Management Strategy : Risk Management Strategy Objective Function Absolute Risk (Less variation) Relative Risk (Pricing variations) Return Underlying Portfolio remains intact Dynamic Risk Overlay Carbon Currency Interest Rate Credit Inflation Weights Duration 6

Composition of Funds : Composition of Funds Government Bonds Corporate Bonds Inflation Linked Bonds Emerging Market Bonds Equity Convertibles 7

Composition of Risks : Composition of Risks 8

Hedging Efficiency Varies Across Asset Classes : Hedging Efficiency Varies Across Asset Classes Risks Hedging Instruments Equity - Equity beta risk - FX risk - Equity Index Futures - FX Forwards Corporate Bonds Duration risk Corporate spread risk - Fixed income futures - CDS on index level ( iTraxx , CDX series) Convertible Bonds Duration risk Corporate spread risk Equity risk - Fixed income futures - CDS on index level ( iTraxx , CDX series) - Equity inde x futures Commodities Commodity prices FX risk DJ UBS commodity future FX forwards Sovereign Bonds Duration risk Country spread risk Fixed income futures Inflation swaps Inflation -linked Bonds Duration risk Inflation risk Country spread risk FX forwards (hard currency bonds) US Treasury futures (hard currency bonds) CDS on EM indices (hard currency bonds) Emerging Market Bonds -Country spread risk FX risk Duration risk - Equity index futures (in case of REIt investments) 9

Cost / Valuation / Risks : Cost / Valuation / Risks Cost of hedging (price), liquidity risk, what to do when to do. Volatility smiles 10

Deciding Risk Techniques : Deciding Risk Techniques B ullish , neutral or bearish Over 150 macro alpha on sources Yield curves and forward rates Correlation will always change in the time of crisis and boom 11

Interest Rate Models (MATLAB) [21] : Interest Rate Models (MATLAB) [21] There are around 10-15 models for interest rate dynamics. Vasicek Model for interest rate to revert to long term mean (FRM P2 Market Risk). How to find the hedge that needs to cover interest rate effect of all components In Derivative toolbox MATLAB uses to create interest rate Trees: Black- Derman -Toy (BDT) Modeling . Heath- Jarrow -Morton (HJM) Modeling Readings: Pricing interest rate swaps from (CFA L2 Derivatives) See Derivatives MATLAB class & Fixed Income for MATLAB 12

Interest Rate Payer Swap / Swap Options : Interest Rate Payer Swap / Swap Options Get interest rate paths based on appropriate model Pre defined models that can be used but needs to be calibrated to include three phases of markets Interest rate objects and portfolios in MATLAB 13

BBB risk management : BBB risk management Index of BBB bonds Tranches protection CDS Highest yield 14

CDX Index Options [7,24,25] : CDX Index Options [7,24,25] Probability of default by models such as Merton Model from market equity price. Other Default models like ratings given by Rating agency. 15

Variance Swaps [8] : Variance Swaps [8] Not much content, and still remains an area of research to implement on MATLAB 16

Options on VIX [9 , 16,17,19] : Options on VIX [9 , 16,17,19] VIX option and future using MATLAB: Implementation not available but can be worked out on requests Mean reverting models 17

Importance of Runtime Analysis : Importance of Runtime Analysis Important steps to keep things working. Human check. Runtime analysis of Risk modules and recombine for efficiency purposes. 18

Optimization Problem : O ptimization Problem Step 1 : theoretical background and formulation of optimization problem. Step 2 : create version of Risk Profile that displays relative values (relative price movements, relative risk data ). Step 3: Optimization based on costing of various hedges vs VAR. Also see optimization toolbox MATLAB class 19

Portfolio Risk : Portfolio Risk VAR calculations CVAR MATLAB [21] Also see FRM L2 Class 20

Liquidity Risk : Liquidity Risk FRM L2 LD= Q/(.1*V) Also see FRM L2 Class 21

Currency Options [12] : Currency Options [12] Check this- seems like not in CFA L2 or FRM L2 Currency Overlay Volatility smiles Currency Indices (Harvest, Forward bias) 22

Equity Options : Equity Options How to move beyond theories Pricing on MATLAB Dynamic Hedging [10] Delta Neutral Hedging Short termed approach Short straddle index can be used [that is how options are created] 23

Carbon Risk [13] : Carbon Risk [13] Quant valuation – not so much 24

Inflation Risk [4] : Inflation Risk [4] Pricing inflation linked bonds Interest rate and CPI behaving in the same way 25

Strategic allocation / Risk adjusted returns / VAR Methods : Strategic allocation / Risk adjusted returns / VAR Methods Summing VAR of various sectors VAR methods as taken from banks Different VAR for different times 26

Distributions (FRM Market Risk) : Distributions (FRM Market Risk) QQ plot Correlation copulas Why normal fails / Other distributions to use? Regime change Test them back FRM L2 books 27

Copula / Correlation [14] : Copula / Correlation [14] Correlation fails Distributions change 28

Implementation : Implementation Step 1 Decide risk budget VAR for a strategy and not for a portfolio Expected shortfall should be the risk budget as that is the loss beyond VAR Step 2: Develop and back test Two possible scenarios: Implemented with market is high and vix is high meaning a change is eminent but this has to be done using future Just before crash buy protection as VAR would be breached 29

SQL in MATLAB : SQL in MATLAB MATLAB using centralized SQL data (that are uploaded and downloaded into and from a centrally hosted server like Yahoo Bloomberg data. Integration of SQL with MATLAB For more details see MATLAB Database Class 30

MATLAB Toolboxes Used : MATLAB Toolboxes Used MATLAB Optimization toolbox Matlab (including use of toolboxes ) Der pricing/ interest rate pricing/ fixed income pricing etc. 31

C# for Financial Engineering [18] : C# for Financial Engineering [18] Basic idea of why we use C# C# for financial engineering book 32

Bloomberg API [11] : Bloomberg API [11] The established service provides free, unrestricted access to raw data for customers for its financial market information. The same publish/subscribe and request/response interactions available via its proprietary interface can be accessed via API. This functionality gives access to data on current market trades, either real-time or delayed, along with reference data on reference data, historical information, and records of intraday trading. API methods support selection by security using standard ticker symbols and the provider's own "Open Symbology " across classes of securities. Effective dates and date ranges may also be specified to retrieve historical results. Sources of pricing data are designated. 33

Summary : Summary Getting risk managed for the fund automatically in a systematic but passive way. Modeling pricing of protection, VAR and the scenarios 34

References : References http :// https :// media/OverlayStrategies_sp.pdf media/Alexander_Preininger-Managing_risk_and_hedging_inflation.pdf https :// media/Overlay_management-Hedging_risks_while_preserving_alpha.pdf http:// 35

References (Contd…) : References ( Contd …) 36

References (Contd…) : References ( Contd …) 37 Wiley: C# for Financial Markets - Daniel J. Duffy, Andrea Germani


Shivgan Joshi
Free Class on: CFA,BAT,FRM, GMAT, Quant FinanFollow me to get updates.
1805 Members Recommend

Your Facebook Friends on WizIQ