Serving the Quantitative Finance Community

 
Michael_E
Topic Author
Posts: 58
Joined: July 11th, 2016, 12:49 am

How to adjust an optimized covered long short portfolio

January 21st, 2017, 10:01 pm

My ambitious model or project study I need some advice in building for study and implement on fly..
BUILDING A WORKABLE MODEL ADVICE

The test is to decide on how a portfolio below compares to various other strategies, passive, buy & hold portfolios --- and see which yields better returns over long periods. 

PORTFOLIO GOAL:
  • Maximize overall returns above passive buy & holds of several types, and benchmarks
  • To maintain by years end a lower annualized portfolio volatility (std deviation) than all above mentioned alternatives
  • Use of leverage will be minimized and will never be 100% weight invested in a market portfolio, but will fix broken positions with cash balances on a short term basis
  • the idea is to beat buy & hold strategies of various kinds, beat the index, with maximized returns over all them while maintaining a  lower annualized portfolio STD Deviation than all aforementioned.
  • ALL FEES, transacation costs, and taxes will be adjusted by years end and compared.
KEY ASSUMPTIONS
  • Not a mutual fund, so no need to buy back shares daily. 
  • Closed ended fund with lockout for a year from initial start
  • fixed amount of shares.   No reissues
  • No bonds, except tactical positions in bond futures
  • No credit derivatives
  • Any Cash not in the market portfolio will be in liquid money market (reasons needing liquid cash in emergency scenarios leg downs)
  • Transaction fees do apply
  • Slippage in early exercises are only accounted for when re-entry to a position at another share price (fees & costs of re-entry accounted for)
  • Some positions called will not seek re-entry, executive decisiveness used when looking for arbitrage on volatility plays and moves in short term stock;
  • On Margin without T-3 regulation
  • Years end will adjust for all taxes for all observed portfolios & will be considered in the return on investments annualized
PERFORMANCE ANALYSIS CRITERIA NEEDS
 daily, weekly, monthly, quarterly, yearly comparisons to similar buy and hold strategy against your portfolio
and plotting the returns against buy & hold, against index, and compare all risk metric ratios with charting of efficient frontiers of your active portfolio strategy against various different Buy & Hold portfolios resembling your tactical covered portfolio (described below).   


TACTICAL ASSET COVERED PORTFOLIO CONSTRAINTS & CONSIDERATIONS
  • You have an optimized portfolio model that allows short sales
  •  You're using a model that may seem like Black Litterman in that it considers excess risk premia that you use intuition to forecast daily.
  • you account for shrinkage.  in the model
  • You are a  selling only covered short options positions
  • If you are short XYZ stock you would only sell a put that covers the position on a high volatility day and you're short vol
  • if you are Long ABC stock you would only sell calls if you expect a shortfall or see extreme jumps up and you sell the high vol calls against it near the money
  • You are allocated with a cash cushion of 40% to 20% at all times just in case you need to quickly short futures to hedge
  • You want to be tactical --- and want to take advantage of short term moves by observing jumps and selling high options premiums  against a stock position long or short the stock with a covered put or covered call
  • You don't necessarily care if you get early exercise on your losing options positions --- but need to account for SLIPPAGE in those moments if you want to get back into a new position
  •  You want to model it with transaction fees on a % scale. 
  • You'd like to adjust your efficent frontier with slippage of covered positions (early exercise) and rolling active futures against it.
  • You want to rarely if ever buy back short options against covered stock
  • you want to be BETA HEDGING to a benchmark (Index, commodity, etc) or multiple benchmarks with a 60 day time correlation
  • You'd like to calculate the sharpe ratio and drawdowns and max values at risk with active options positoins and their premiums & greeks
  •  You are not always going to be covering positions with options.  With low vol.  not always going to be expensive enough to justify selling premium on the positions.    You would wait for the oppurtunity when it might come.   it might not come
  • you'd like to know all your costs of slippage early exercises and the cost of transactions and the price to re-enter the position later.
  • you'd like to maximize dividend collections and keep the portfolio running without taking profit too early
  • you want to close out losing bets quickly but not too quickly!  
  •  how to determine when to exit a losing trade?  
  •  You would only be trading around active liquid options market for the underlying, no giant bid:ask spreads just tight and liquid as possible underlyings
  • continuously hedging only covered short premium options against those long & short  near the money premiums on the fly on high volatility action that would be on a day when you might find 350% change in liquid options at near 25 delta or even 50 delta and doing nothing on other days
  • You want to measure parameters of risk at end of day discretely quickly and account for all above scenarios
    MORE THAN AMBITIOUS.
    • USE of parrallel GPU computing
    • Compiled executable GUI dashboard access with backend SQL access
    • a GUI with a dashboard, gauges, charts, and benchmarks against it on live tick data
    •  A dashboard that automatically adjusts continuous risk metrics on tick data such as rolling parametic VaR, CVaR, Jensen,  Sharpe,Sortino,  based on all your trading activity as the input.
 
Michael_E
Topic Author
Posts: 58
Joined: July 11th, 2016, 12:49 am

Re: How to adjust an optimized covered long short portfolio

January 22nd, 2017, 12:34 am

THE ADDITIONAL CONSTRAINTS & LOGIC BEHIND IT
  1.  thorough management of options greeks from the selller's edge.
  2.  No naked premium selling, but partially covered positions or fully uncovered positions to continuously tightly manage portfolio DELTA
  3. We wouldn't want to stay neutral and collect premiums alone, but we'd like sell high premiums, be semi directional with tight deltas  being long or short correlation on other assets for 60 days which may be covered with short premium or just beta hedged to a liquid futures contract that would be small in size, liquid and easily bought and sold to hedge out any kinks
  4. WHEN THE VIX is at all time lows, and the market is completely complacent they're positional in some direction while I'm neutral or slightly directional collecting premiums in the meantime, and if I decide to get directional..  I'll adjust my deltas by either selling or buying more stocks..  but the idea is to be determinant  with flexibility and hedges --- be liquid and sometimes just taking on positions hoping to expire DEEP ITM on short premium, because I don't want to hold the stock beyond expiry --- I just wanted to collected the premium without risking naked margin on it,  
  5. If I want to switch my position and be long or short -- I can parameterize it through the deltas which were managed through highly uncorrelated assets against the short premium positions i collected all the premium from and was saved the trouble of having to sell the asset I didn't want to hold --- i just wanted the 400% premium and hedged with correlation against it..  and used deltas to manage the risk (according to additional deltas either positive or negative against any position i'm
  6. BE DETERMINANT when you want to --- only by adding extra bits of delta or short deltas when you want to --- but the idea is to collect highest premiums without be market directional at times, and other times?  being determinant with tightly correlated betas and by increasing deltas in our directional bet --- which is the primary idea.
  7.  no naked premium selling, but delta management is done by correlation and beta hedges with other liquid futures contracts
 
Michael_E
Topic Author
Posts: 58
Joined: July 11th, 2016, 12:49 am

Re: How to adjust an optimized covered long short portfolio

January 22nd, 2017, 1:12 am

i MUST BE STUPID. --but i've been doing and doubling portfolio value year on year for 6 years and with a lower portfolio standard deviation than all the major indexes i'm hedging against or with..

and I must be an idiot..  because it just keeps growing in value year on year..  by 1 = 2 by years end with transaction fees and taxes ---  and it just keeps doubling and i'm never usually fuly bulish or bearish on anything except i'm playing arb and collecting no matter --- and hedging around that angle in real time with all the other assets... so i'm collecting in sideways markets and directional markets I simply just add more delta --- I'm in and out quick because i've correlated tightly to futures positions against it already
 
Michael_E
Topic Author
Posts: 58
Joined: July 11th, 2016, 12:49 am

Re: How to adjust an optimized covered long short portfolio

January 24th, 2017, 6:34 pm

i DECIDED..  nah.   I don't want to use this strategy.   All the ITM calls and ITM puts got called..  I took profits..   didn't go and buy or sell much at all except LONG GAMMA PUTS and CALLS..

What's my sharpe ratio?
 
Michael_E
Topic Author
Posts: 58
Joined: July 11th, 2016, 12:49 am

Re: How to adjust an optimized covered long short portfolio

January 24th, 2017, 11:52 pm

SPENT 100 hours in last 8 days working in R studio.   Only to realize that I work too damn slow to use any of my analysis in the markets. 

Also, that we live in a bizarre age where speed is so necessary to combat changing conditions,
no longer am I believer in a static way to manage ---  

EXCEL or end of day data  --- NO GOOD!

HOWEVER
WE Can't go in there blind!
we need some ratios   and  there's just to much data to fetch off the web scraper...   then to parse and fit into data frames for one person to realize before it's too late how screwed they could be soon!

  1. ..  we need a way to LEVERAGE PROCESSING POOLS of PARALLEL computers --- some locally just storing data files in indexed and neatly managed vector/matrixes of equal lengths..
  2. then we need a vertical POOL of parralel processors calculating actual NYSE /NASDAQ/CBOT/CBOE data into the environment directly on a TICK basis..  to calculate metrics on a dashboard...
  3. SO --- you know your exact data every second, and if you're making trades --- it should update those metrics.    It's a big project.
PASSIVE MANAGEMENT
has the advantage that when markets tank, they all fail together and no one is an idiot.


WE NEED THE RATIOS QUICKLY on tick basis --- life is changing fast.
this market moves so quickly.   You can't keep up with it.

TRUMP - certainly stirring it up.
Anyone know what he'll do next?
RAISE RATES?   NAHHHH....   that's another rant from him
!
Buying bonds?    nah..  can't
Selling bonds?    nah.. can't
Long the market?
no.   kind of just sitting in cash waiting for it to make sense.   but i suspect it could burst upward or collapse downward in a matter of seconds. quiet before the storm..   LOW VIX ----- buy a straddle or two?
a few open puts and calls.. 


OTHERWISE..  it's useless by the time you've calculated your portfolio.   ----  the market crashed on you!

Image
YIKES

Image

Image

SP500 ROLLING RETURNS  --12 month
Image
 
Michael_E
Topic Author
Posts: 58
Joined: July 11th, 2016, 12:49 am

Re: How to adjust an optimized covered long short portfolio

January 24th, 2017, 11:57 pm

FML..  I need a Parrellel GPU pool to simply parse ZOO and XTS objects for me!

TOO MUCH DATA to analyze..   Not enough time to effectively deploy a strategy.
UNLESS you're a nobel economist,  or two  ---  then  you can be passively long in this market..

LESSONS:  ROBERT MERTON & MYRON SCHOLES -- LONG TERM CAPITAL MGMT
http://www.nytimes.com/2008/09/07/busin ... 41880.html

We are seeing the highest open interest in 20 years on 10 year TREASURIES.

Long?   Nahh.   Can't be  --- Trump is unpredictable and so is the mkt
Short?   Nahhh.   Can't be 

-----------  LEVERAGE CYCLES -----------
with all this open interest on BOND FUTURES..   you'd expect to see some people selling off other assets in their flee from whatever positions they were holding to liquidate highly levered positions when the margin requirements roll over them! 

I say...   Stick to straddles ---  but get the hell out of the way---  it's either up or down and volatility in both directions..   

but today?  VIX is low... 

--------------------------
SELL IT ALL!     and buy straddles

then figure out how to fix the static portfolio management theory,