wilmott.com

Thanks to greghm and jamesnowak.P

A mishedged short gamma asset allocation strategy with an unknown non-linear payoff.The "control" on the random process is dreamt up by structurers - not that they think of it in this way. It does sell though.

CPPI = Constant Proportion Portfolio insurance.Very simplified: Suppose you have an investment horizon of 5 years and 100 $ at the moment. Buy a ZCB (Zero Coupon Bond) today for say 70$. With the other 30$ you can now do more risky things. Worst case: After 5 years you still have the 100$, you lost the risky investment! And of course you lost the interest on the 100$ - There is no free lunch.The more complicated part about the CPPI: the whole thing is done more dynamically with some adjustements... if interested, I have a simple R-script which simulates it with a GBM (Geom.Brownian Motion) for the risky and a constant interest for the riskfree asset... but there are more sophisticated programms out there.soliton

Portfolio insurance, Alternative method for guaranteeing capitalDynamic allocation using a dynamic rebalancing between a Risky and a Riskless Asset depending on the Underlying (“U/L”) performance If the U/L does not perform, the CPPI decreases its exposure.The main risk of CPPI to the CPPI Manager (Hedger) is the Gap RiskGap risk is the risk of a serious downfall of the underlying beyond the limit allowed to provide the capital back at maturity A few parameters :The Bond Floor : calculated at inception, it’s the Zero Coupon level corresponding to the maturity of the CPPIMultiplier : It calculates the exposure. Multiplier = (1 / Gap Size)Example : If Gap risk = 20%, Multiplier = 5Portfolio value : Sum of the risky and riskless assetThreshold : Limits inside which, the CPPI Manager assesses that there is no need to re-balance. More commonly used for CPPIs on illiquid products (Funds, Credit Derivatives structures) since it’s all linked on liquidity of the U/LA few terms : Cushion or Health = Portfolio Value – Bond FloorThe only subjective data for a CPPI Manager are :Gap Size & Fees. Management Fees acts like a dividend, they reduce the strategy price

Usually structured as a combination of three parts: risky asset unit, zero bond unit, and the leverage unit.When the risky unit is invested in illiquid assets, such as hedge funds, it becomes tricky to estimate the Gap risks.CPPI is a negative gamma product. Wonder why investors want it at all.

People always say it is a negative gamma structure, because you buy high and sell low.But it is also known that the vanilla CCPI structure resembles a power option - which must by nature be long gamma?See for instance CPPI power option paper, esp. graph 4A.Does anyone have any thoughts on this?

Hi I would really like to have a look at your file, if its not a problem for you...thanks

QuoteOriginally posted by: solitonCPPI = Constant Proportion Portfolio insurance.Very simplified: Suppose you have an investment horizon of 5 years and 100 $ at the moment. Buy a ZCB (Zero Coupon Bond) today for say 70$. With the other 30$ you can now do more risky things. Worst case: After 5 years you still have the 100$, you lost the risky investment! And of course you lost the interest on the 100$ - There is no free lunch.The more complicated part about the CPPI: the whole thing is done more dynamically with some adjustements... if interested, I have a simple R-script which simulates it with a GBM (Geom.Brownian Motion) for the risky and a constant interest for the riskfree asset... but there are more sophisticated programms out there.solitonHi,u mind if send the R-script to me..that will be really help for me..thanks..

Good summaryhttp://en.wikipedia.org/wiki/Constant_proportion_portfolio_insurance

Hi all,Any good articles or book I can read on CPPI as the company I am is doing these stuff now

I would be interested in the R script. Would appreciate if you could send it at jaina2@gmail.com

Hi soliton,i like to see your script if you don't mind to send it to me. CheersF

python script here if anybody wants to have a look ... used it for prototyping so not sure if its all bulletproofWould recommend using it in the ipython application ... can plot averaged outputs using ->plot(numpy.average(CPPI,1))Python CPPI Scriptthought folks might be interested

I'm thinking: from CPPI buyer's perspective, gamma is positive (as graphed in the Boulier's paper). However, from seller's (institutional) perspective, gamma is negative. Analogously, if bank sold a vanilla call, buyer gets positive gamma while bank gets negative gamma. From buy high sell low perspective: a short call position (bank), when spot moves from OTM to ITM, the delta increases, so to hedge this short position, banks also need to buy more delta when spot moves higher.