9,349 reputation
1530
bio website
location Hong Kong
age
visits member for 2 years, 9 months
seen 8 hours ago

Nov
7
comment Implied volatility and pricing of vanilla options
...be able to extract risk adjusted value. I do not have more to add to this except maybe mentioning that BS does not itself prescribe how you are to hedge an option position but its derivatives do prescribe how to hedge specific risks.
Nov
7
comment Implied volatility and pricing of vanilla options
I believe that implied volatility of any basic asset has to be formulated. Whether you agree with market consensus or whether you build your own forecasting model. The same applies to exotic products with complexity of your choice. The volatility models perused there depend on the very same basic building blocks such as a vanilla option and its implied volatility. So, even stochastic volatility models rely on the same basic mechanics than your own implied volatility forecasting model. Volatility is not deterministic and hence anyone who can model future volatility better than the market will
Nov
7
comment Implied volatility and pricing of vanilla options
Forget the "BS world". BS is a translation tool, it means next to nothing. It is a market agreed tool to statically express volatility in terms of currency denominated price. Hence it is completely irrelevant whether BS exhibits flaws or not as long as all market practitioners agree on the usage of the exact same tool (which they do, at least in the equity world). BS simply translates your volatility figure into a tradable price at inception of the trade. Nothing else. What the market is trading is volatility not option prices.
Nov
7
comment Implied volatility and pricing of vanilla options
Partially, instead of occupying your time to predict future option prices you can "simplify" by predicting the implied volatility. Not that it is any easier but you somewhat boil it down to the essential. Some option traders believe they are more successful at modeling volatility than modeling asset prices. Hence they hedge out other risk exposures and focus on volatility trading. Btw, other BS inputs are anything but static over time. Consider the underlying price, consider even something as "trivial" as dividend curves of single name equity options.
Nov
7
comment Implied volatility and pricing of vanilla options
As Mark Joshi pointed out your questions seem philosophical. Vanilla option prices are nothing more than a reflection of the market's take on implied volatility. How you arrive at such volatility estimate is entirely up to you. If you believe you have a superior model to arrive at implied volatilities (aka, if you think you are able to better predict future price variation of the underlying) then employ whatever you like and trade it against market prices. You should over time extract alpha if your model is indeed superior.
Nov
3
comment Automatic trading strategies - what are benchmarks for PL on serious backtesting?
I do not mean to be condescending, I just share what I think the simple truth is. You need to figure out for yourself which asset class over which holding periods suits you best and whether you attempt a momentum or mean-reverting type of approach or whether your game is in arbitraging inefficiencies. There is no road map, it starts with your skill set, experience and what style suits you the best. Even if the holy grail existed (which it does not) and was given to someone it would most likely be worthless in such person's hands. I hope you see where I try to get to...
Nov
3
comment Automatic trading strategies - what are benchmarks for PL on serious backtesting?
a) if it is not difficult to backtest simple strategies then why don't you do it? Indeed it is not difficult. So not sure what is keeping you rather than "spending many hours searching for results". (Each strategy and each tested data set produces different results hence your quest for an average performance makes very little sense imho). b) please feel free to aggregate the performance results of the many hundred hedge fund reports you can get access to; you will most likely spend months with little to show for. c) I test my own strategies and do not intend to share results. Sorry
Nov
3
comment Automatic trading strategies - what are benchmarks for PL on serious backtesting?
I still do not fully understand what you try to ask. Obviously there are strategies in the market that generate alpha, most do not. Like with everything in life the probability of success is a function of skill, hard work, adaptability, and some luck. There are obviously no statistics out in public domain that quantify the average return of trading algorithms in existence/operation...
Nov
2
comment Automatic trading strategies - what are benchmarks for PL on serious backtesting?
Your question is way too broad. Please rephrase the question if you can. Also many of your terminology is not quantifiable. What is "serious" back-testing?
Oct
29
comment Where to find stock buybacks yields?
what is a stock buyback yield? There are various types a company can repurchase shares? In the open market, through private negotiation, among others. Thus, not always is a premium paid on top of fair value.
Oct
27
comment How is stock data objectively different to this random walk?
That is not what EMH states (that future prices follow a random walk). Even the statement in Wiki is plain wrong: "This implies that future price movements are determined entirely by information not contained in the price series. Hence, prices must follow a random walk." Just because future prices may not be a function of past prices does not infer future prices to follow a random walk.
Oct
27
comment How is stock data objectively different to this random walk?
You look at 250 data points. If you look at a longer term time frame you will notice the absence of upward drift in your random data model vs the drift component that partly drives equity (and its index) returns. For higher frequencies the quantity of "white noise" increases and a portion of any researcher's job becomes to apply suitable filters in order to isolate the time spans when non-random pricing data allow for alpha extraction.
Oct
27
comment Build a customizable trading engine in python
Way too broad question. You emphasize customization requirements and that basically requires an incredible code stack that will be beyond any simple project. Just want you to be aware of the fact that you will most likely spend a year if you build from scratch. And I highly recommend not to attempt this project in Python but an OOP language that is much more performant, with a more stable and mature code base and that is actually suited to handle modules like broker API connectivity, OMS, PMS, parallel event based processing, and the like. Just my 2 cents...
Oct
22
reviewed No Action Needed Negative Eonia rates
Oct
22
comment How do I calculate Sharpe ratio from P&L?
@afekz, well in the end you choose your tools. Anyone who uses Sharpe ratio metrics should know precisely what they measure. You cant come later and complain that SR did not account for operational risk or model risk. And I simply pointed out that SRs can be derived from P&L, regardless of the avg holding period of positions. And I am not happy to digress because I prefer to stay on topic. If you want to explore better measures then please open a new thread or ask a different question.
Oct
21
comment how to use known premium of options to determine premium of options with another strike?
Thanks for the clarification. I would still disagree with the described "shape". Especially currency option volatility smiles can hardly be described as linear on either side of the atm strike.
Oct
21
comment Where can I find literature (books, articles, etc.) about basic HFT / arbitrage strategies?
Excuse me but how are ANY of the above topics related to high frequency trading? Most everything in hft is related to the efficiency of hardware and technology rather than making probabilistic assumptions and/or predictions (other than of the most simple and computationally efficient kind). You would never reach microsecond space if you had to perform calculations that originated from any of the above topics you listed. And please do not confuse the latency space Simons operated/operated in with hft.(-1)
Oct
21
comment how to use known premium of options to determine premium of options with another strike?
@barrycarter, could you please elaborate? It is either linear or exhibits different slopes on either side (in which case it is hardly linear). I would argue that the function of the smile highly depends on the exact type of underlying asset but rarely is it linear (except in special cases).
Oct
18
comment How do I calculate Sharpe ratio from P&L?
I agree with you that Sharpe is a poor metric, there are better ways to assess risk adjusted returns. But certainly it is not to model the probability of an operational blowup. Its like heavily overweighting the jump probabilities in a jump diffusion model as part of option pricing, for example. And of course does a lot of the intraday volatility (hopefully) vanish when the law of larger numbers kicks in, that is precisely the central reason for the existence of hft shops ;-)
Oct
18
comment How do I calculate Sharpe ratio from P&L?
@afekz, I disagree. Of course does anyone in this space primarily concern himself with trading risk not operational risk. You can't argue that just because every now and then an operational issue blows up everyone thinks all day long about how much it would cost if their servers or some tiny bit of code blew up. Yes maybe an operational manager. But firms like this are run and driven by those who generate revenues not those who keep costs down. Profit centers have always trumped cost centers.