Читать книгу Positional Option Trading - Euan Sinclair - Страница 10
ОглавлениеINTRODUCTION
You know nothing, Jon Snow.
—Ygritte in A Storm of Swords by George R. R. Martin.
He is not the only one.
We are not in a time where reason is valued. In economics, the idea that marginal tax cuts pay for themselves is still advanced, even though all evidence says they don't.
Forty percent of Americans do not believe in evolution. Forty-five percent believe in ghosts. These beliefs are not based on any evidence. They are manifestations of another philosophy, whether it is economic, religious, or sociological. Usually these opinions reveal more about what people want to be true rather than any facts that they know. And many people know few facts anyway. Evidence is seen as irrelevant and arguments are won by those who shout loudest and have the best media skills.
The idea that opinions are as valid as facts also affects trading and investing. Many investors rely on methods that are either unproven or even proven to be ineffective. The few of these investors who keep records will see that they are failing but rely on cognitive dissonance to continue to believe in their theories of how the markets behave. One would think that losing money would prompt reexamination, but the persistence of losers is remarkable. And even when some people give up or are forced out, there is always new money and new participants to replace the old.
The only way to learn about anything is through the scientific method. This is an iterative procedure where theory is modified by evidence. Without evidence we are just in the realm of opinion. Most of what I present here is backed by evidence. There are also some opinions. My justification for this is that experience is also a real thing. But I'm no less prone to self-delusion than anyone else, so feel free to pay less attention to these ideas.
Trading is fundamentally an exercise in managing ignorance. Our ability to judge whether a situation presents a good opportunity will always be based on only a simplified view of the world, and it is impossible to know the effects of the simplification. Our pricing model will be similarly compromised. It will be a simplification and possibly a very unrealistic simplification. Finally, the parameters the models need will have estimation errors and we generally won't know how large these are.
It is impossible to understand the world if you insist on thinking in absolute terms. The world is not black and white. Everything has shades of gray. You won't learn much from this book is you aren't comfortable with this.
This is clear for “risk.” Everyone has a different risk tolerance, whether this is personal or imposed by management or investors. But more important, risk is multidimensional. We are comfortable with this idea in some areas of life. Imagine you have a choice of going on vacation to either Costa Rica or Paris. Both are nice places and any given person could reasonably choose either one. But Paris has no beaches and Costa Rica doesn't have the Pompidou Centre. There is no one correct choice in this situation. And that is also the case in most investing decisions.
Many of the ideas I write about can be extrapolated to a ludicrous level. But if you do, don't blame me or credit me with the resulting conclusions.
In no particular order here are some facts that are often misrepresented:
There is usually a variance premium. This does not mean there is always a variance premium.
There is usually a variance premium. This does not mean you should always be short volatility.
Short volatility can be risky. This does not mean that short volatility has to have unlimited risk.
Some theories (e.g., GARCH, BSM, EMH, returns are normally distributed) have limitations. This does not mean the theories are stupid or useless.
The Kelly criterion maximizes expected growth rate. This does not mean you should always invest according to it.
If what I write is unclear or incorrect, that is a problem of my making. But if you choose to ignore nuance, that is your issue.
Trading as a Process
I have made no attempt here to write a comprehensive option trading book. I don't cover the definitions and specifications of various types of options. There are no derivations of option pricing models. I expect the reader to know about the common option structures such as straddles, spreads, and strangles. Many books cover these topics (e.g., Sinclair, 2010). A very brief summary of the theory of volatility trading is provided in Chapter One, but this is not a book for beginners.
This is a book for experienced traders who want the benefits of including options in their strategies and portfolios but who are unwilling or unable to perform high-frequency, low-cost dynamic hedging. Again, there are many books on this type of positional option trading, but none are theoretically rigorous, and most ignore the most important part of trading anything: having an edge.
One of the things that distinguishes professionals from amateurs in any field is that professionals use a consistent process. Trading should be a process: find a situation with edge, structure a trade, then control the risk. This book documents these steps.
The book's first section explores how to find trades with positive expectation.
In Chapter Two, we look at the efficient market hypothesis and show that the idea leaves plenty of room for the discovery of profitable strategies. This insight lets us categorize these “anomalies” as either inefficiencies or risk premia. These will behave differently and should be traded differently. Next, we briefly review how behavioral psychology can help us and also its limitations. We examine two popular methodologies for finding edges: technical analysis and fundamental analysis.
Chapter Three looks at the general problem of forecasting. No matter what they say, every successful trader forecasts. The forecast may not be one that predicts a particular point value, but probabilistic forecasting is still forecasting. We introduce a classification of forecasting methods. Forecasts are either model based, relying on a generally applicable model, or situational, taking advantage of what happens in specific events. We very briefly look at predicting volatility with time-series models before moving on to our focus: finding specific situations that have edge.
The most important empirical fact that an option trader needs to know is that implied volatility is usually overpriced. This phenomenon is called the variance premium (or the volatility premium). Chapter Four summarizes the variance premium in indices, stocks, commodities, volatility indices, and bonds. We also present reasons for its existence.
Having established the primacy of the variance premium, Chapter Five gives eleven specific phenomena that can be profitably traded. The observation is summarized, the evidence and reasons for the effect are given, and a structure for trading the idea is suggested.
The second section examines the distributional properties of some option structures that can be used to monetize the edge we have found. We need to have an idea of what to expect. It is quite possible to be right with our volatility forecasts and still lose money. When we hedge, we become exposed to path dependency of the underlying. It matters if a stock move occurs close to the strike when we have gamma or away from a strike when we have none. If we don't hedge, we are exposed to only the terminal stock price, but we can still successfully forecast volatility and lose because of an unanticipated drift. Or we can successfully forecast the return and lose because of unanticipated volatility.
Chapter Six discusses volatility trading structures. We look at the P/L distributions of straddles, strangles, butterflies, and condors, and how to choose strikes and expirations.
In Chapter Seven we look at trading options directionally. First, we extend the BSM model to incorporate our views on both the volatility and return of the underlying. This enables us to consistently compare strikes on the basis of a number of risk measures, including average return, probability of profit, and the generalized Sharpe ratio. Chapter Eight examines the P/L distributions of common directional option structures.
The final section is about risk. Good risk control can't make money. Trading first needs edge. However, bad risk management will lead to losses.
Chapter Nine discusses trade sizing, specifically the Kelly criterion. The standard formulation is extended to allow for parameter estimation uncertainty, skewness of returns, and the incorporation of a stop level in the account.
The most dangerous risks are not related to price movement. The most dangerous risks are in the realm of the unknowable. Obviously, it is impossible to predict these, but Chapter Ten explores some historical examples. We don't know when these will happen again, but it is certain that they will. There is no excuse for blowing up due to repeat of a historical event.
It is inevitable that you will be wrong at times. The most dangerous thing is to forget this.
Summary
Find a robust source of edge that is backed by empirical evidence and convincing reasons for its existence.
Choose the appropriate option structure to monetize the edge.
Size the position appropriately.
Always be aware of how much you don't know.