. . . There is one aspect that I would like to know as it influences in trading decisions and are fundamentals.........
here is my 2c based on my 26 years of trading experience..... ..
)Here is how I would approach it.
There are two broad styles to harvesting alpha in the market.
- A specialists discretionary approach of expert unique knowledge and logic applied to a few select investments - This is where I would classify the FA guys (eg. global macro) and the Value Investors (aka the Buffet's);
- A generalists approach of systematic application using the same underlying logic associated with a general market principle applied to a wide universe of investments (eg. mean reversion and momentum).
I have a feeling that you may be getting a little pregnant with Approach 1 which does compromise the philosophy of Approach 2. Both approaches are valid....yet you do need to take a stance on which approach suits you better......otherwise you will find that your performance metrics grossly alter.
Each have their pluses and minuses but they require different skill sets and risk tolerances.
Approach 1 is for the engineers who apply the principles of 'cause and effect' to market action. They hate uncertainty and like precision. They want a rational reason for why things move. They tend to view the markets as wrong when things don't work out the way it is planned.
Approach 2 is for those who view markets as complex moving feasts. More for quants and mathematicians. They use statistics to address market uncertainty. They accept many losses and never look at a simple trade in isolation. They look at the next 1000 trades as a sign of success. They always view the markets as right.
Approach 1 takes an intense market research point of view where the aim is to gather market/investment information that is not available to the general participant. In a nutshell these guys are seeking to exploit information that is not available to the general participant to gain an arbitrage opportunity. These guys are typically expert analysts who have a thorough understanding of the investments they trade and can determine with a high degree of precision the current and future intrinsic value of an instrument. They can then assess whether:
- current price is above/below the intrinsic value and take a position to capitalise on the future prosensity of price to converge towards that intrinsic value (a convergent approach to trading); or
- adopt a divergent approach to trading if they expect the current intrinsic value of the instrument price to diverge to a future higher/lower intrinsic value.
I have lots of time and respect for these guys but the skill sets required to be good in this game are time intensive and very demanding. Given the extent of analysis that needs to be undertaken, these guys simply do not have the resources to invest their efforts towards a wide universe of investments and hence they are very selective in the instruments they trade and are intensely discretionary in nature.
They have a far greater risk tolerance than my appetite as they have high conviction in their analysis and are willing to bear untolerable drawdowns in their assumption that they are right. The volatility of retrurns of approach 1 however is not my cup of tea. While you may be right in your assessment of future direction, it is the timing that can kill you.....hence the drawdowns while waiting for your analysis to be confirmed. Frequently these guys therefore add TA to apply to their entry and exit decisions. It is the timing issue that causes Approach 1 to significantly vary in performance metrics. You can be wildly successful or you can end up in tears (eg. Valeant Pharmaceuticals). If you plot CAGR against drawdown of FA funds and deep value investors, the scatter plot is all over the place. This is because in this field you can either be wildly successful or horribly unsuccessful dependent on the validity of the info you have obtained. To have consistent sustainable investment earnings from this approach is an exception as opposed to a rule.
Approach 2 takes a more generalist statistical point of view to capturing alpha. The assumption is that each instrument on their own may offer small degrees of alpha applying a simple yet robust approach of a general market principle such as momentum or mean reversion, and thus it is necessary to spread the approach as wide as possible to increase trade frequency and capture a large number of return distributions to compensate for their individual low contribution to overall equity. Because you are applying a simple principle across a broad range of instruments again and again......a scatter plot of CAGR versus drawdown for methods in this approach tend to cluster better. There is less style drift in this approach simply due to the persistent application of the same trading rules. Skill is not as essential in this class of approach...but what is essential is systematic application. If you are not systemised in this game of diversification you are severely handicapped.
You have often heard that EA's simply don't work in all market conditions.....Well that of course is right.....but..... what discretionary traders fail to recognise however is that in the systematic space of this approach you may be trading 10 different EA's (using different style approaches where each EA capitalises on a different market condition) across 4 different timeframes across a universe of say 40 instruments. You simply cannot do this as a discretionary trader...no matter how good you think you are.
Approach 2 is my preferred cup of tea as:
- I am simply not sufficiently skilled to apply FA or value investment to obtain an edge in the markets. There are experts far better than me and research houses with lots of resources at their disposal to outwit my puny efforts. The likes of Hanover, Numbnuts and Sys play in this space and you can see how skilled they are in this game......so imagine what the fund managers with R&D teams at their disposal are like. As Thomas said, general market participants don't have a hope in this area as they are inevitably late to the table in their analysis and the markets are likely to have already factored in these opportunities.
- I am fairly well skilled in system design and application which allows me to beat the general participant by system design and diversification by deploying a large number of strategies across many timeframes and instruments to capture market momentum and crisis alpha. I focus on those low hanging fruit opportunities that simply do not attract the interest of the adrenalin seeking trading junkie or those major 'shock events' where most market participants are in a WTF panic moment. Being diversified allows you to capitalise on market shocks and by following price as opposed to predicting it...... you are more likely to be able to capitalise on crisis alpha.
I therefore apply a probability game to my approach that applies the basis of:
- Trade as many different return distributions as possible to capture in broad scale small amounts of available alpha. A return distribution is generated from every instrument, timeframe and system deployed.
- Having multiple return distributions allows you to significantly benefit from improved risk-weighted returns (through diversification)
- Apply a very small trade risk % to every trade of <=0.50%. I never look at or worry about a single trade....I only worry about the next 500 to 1000 trades. The name of the game for me is survival over a large sample size.
- I never trade for a living. I trade as a wealth creation exercise. You can never predict with confidence any sense of a cashflow. You must have significant capital backing to play this game full-time.
- Most of the trades will be losers so apply stringent risk management techniques that ensure that any loss is miniscule in terms of equity deterioration. Some of the trades, simply due to the Law of Large numbers, will be significant winners. I simply cannot analyse in advance which trade will be successful so I don't even bother. I apply a rinse and repeat procedure on every trade and simply 'follow price action'. Simple as that....no prediction. I therefore do not apply relative strength measures in selection as that is a form of 'predictive selection'.
- Focus on risk management at all times and do not worry about profits. They will simply arise through the Law of Large numbers;
- Don't listen to the news - All my backtesting includes the major news events so as I am confident in my back testing I simply ignore news events and trade through them. Fortunately I trade the longer timeframes so an individual news event tends to be insignificant in the scheme of things. I am after a serial repetetive release of news events that drives momentum in the same direction.
I hope this helps J.
Trade safe C :-)
PS....forgot the musical accompaniment. What was I thinking :-)
