Wednesday 2 May 2007

Bayesian Reasoning + Chart Analysis = Good Stuff

A key, but occasionally overrated and frequently misunderstood, aspect of a market analyst's armoury is trend/chart analysis. Classical economists have depended on such analysis as a basis of their market approach for around a century and the popularity of the creation of charts with supportive analytical conditions still underpins the trading strategies of many market participants.
Thomas Bulkowski, in his excellent book "Encyclopedia of Chart Patterns", manages to isolate over 15,000 different chart formations that may be included in one's analysis of a market sector, currency, stock or bond. Many of these structures are, indeed, robust ie the presence of a particular formation having a high statistical probability of yielding a particular market reaction. For example, many chartists are currently excited by the patterns being exhibited by both the FTSE100 and Wall Street Daily Cash as these patterns are indicative of a market at it's peak and such analysts see in these charts clear evidence of the end of the recent bull run in the global financial markets. Analysts who focus on charts exclusively are likely to create false constructs however and several external influences need to be accommodated in order to develop a complete picture of a market reality.
A naive blinkered belief in charts is illogical. An analyst must also assess the bigger picture. The charts may well indicate a downturn in the global markets but what is the situation on the ground. It is our belief that such an holistic approach is the most critical of inputs to any solution. Supportive of a fall in the major western markets are the following:
* Global warming - The degree of validity to the existence of global warming is irrelevant with respect to a market analyst. Personally, I side with the overwhelming number of scientists who claim that there are significant man-made triggers enhancing the climate change that is evidently in motion. But, from a market perspective, the key factor is psychology. People now BELIEVE that climate change is a reality and this will lead to a market slowdown as governments and businesses are forced to react both on a tokenistic level and structurally. The momentum behind this dynamic is colossal. While we pay lip service to neoclassical dismal science, it is the shared attitude of all our Trading Team that behaviourism is the key mechanism in any financial market. The shared belief of the masses in climate change is a major psychological imprint on the markets.
* Housing Markets - Spain, Ireland, America and Britain have various degrees of bubble within their property markets - in March in the US, sales of existing homes fell by over 8% which is the most precipitous decline in two decades and subprime lending has collapsed while Spain's housing boom is nosediving. In many countries the recent housing hike has funded the recent bull market. Yet property is, in many ways, a fake economy. A British house doubling in price over five years, for example, is not wealth creation. It is merely a transfer of cash from those without property to those with property - a regressive redistribution of wealth, if you like. Building an economic boom around such a frail structure is doomed to end in tears. For some... A lack of savings (particularly in the US) and high levels of personal debt across the west will amplify the impact of the various degrees of stalling in the property markets.
* Taking Advantage of the Mugs - The final splurge in any bull run is when mainstream capitalism targets the financially bewildered. The creation of a system of living that is founded on an almost autistic structure of numeracy which offers a life competitive advantage to some is, by it's very structure, going to disenfranchise huge swathes of the population. Behaviourism raises it's head again here and market amateurs stream into the markets just before they turn. The subprime housing markets in both Britain and the US together with the stockmarket frenzy in the volatile Chinese markets are clear indicators of a major market correction.
* Other Factors - There are numerous other factors impacting negatively on the markets and it is not our intention to address them all here but issues relating to the weak dollar, the failure of the Doha round of trade talks, terrorism, protectionism, the oil market, public perceptions regarding globalisation's wage slavery will all take on the form of a millstone in the coming period.
* Globalisation - the only factor supportive of a continuation of the bull market is globalisation. The transfer of power from labour to capital which is the very foundation of our modern slavery is leading to major global liquidity and the current buying binges of the private equity companies is indicative of a system that has allowed a regressive redistribution to go too far. Although the non-unionised wage slaves of numerous countries are a dynamic towards a continuing market progression, other factors, particularly mass awareness of this slavery and climate change, will undermine this momentum.
Behaviourism is the key dynamic. People do not behave by optimising their own self-interest as implied by the cold "science" of classical economics. People do conform to a range of psychological styles and disorders that may be aggregated to develop a footprint on any market from the FTSE to Liverpool versus Chelsea. Prices are not efficient and markets are irrational. Value may be taken by determination of the true market price compared with the price based on a collective but flawed psychology. Additionally, chaos reigns supreme. Self-similarity and several other equilibrium structures impose themselves on all trend and chart analysis. Chaos and cybernetics form much of the foundation of our trading strategies. The main issue that I have with the utilisation of chart analysis as a trading panacea is that charts are self-fulfilling. We all think that we are very original in our recognition of the factors that impact upon a particular market. As an example, I would guess that I utilise all valid parameters that may affect a football betting market but I would also guess that 95% of these parameters are out there being generally absorbed into the market price. The price consequently over compensates with regard to the original dynamic, yet again exposing value to an analyst.
In an integrated global financial system, the levers of power in the financial sector possess considerably more structural market control than historically. One of the major logics behind the annual Davos shindig is to enable the major global operators to coordinate a strategic template against which all other macroeconomic decisions may be independently or collectively made. In Britain, when Gordon Brown, in his wisdom (sic), decided to hand the Bank of England responsibility for the setting of interest rates, he was effectively giving the financial sector a far greater control of the economy than previously. We could itemise many other examples of this transfer of market control from the executive branch to the financial. A key outcome of this cartelisation of financial control is that markets may be driven to a prescribed agenda. Absolute control is not achievable as, on occasion, the collective global psychology does not behave in an appropriate manner due to complex behavioural feedback loops or computerised trading systems may interact in unforeseen ways or major global breakpoints impact upon the markets eg 9/11, for example. Notwithstanding these factors, a significant degree of control of the markets is exerted by the major global operators. The value of this control for the controllers may be very simply explained by the following example:
If a stockmarket is expected to average, say, an annual return of 9% over a window of 10 years then the market infrastructure could be easily created whereby volatility is dampened and financial bubbles pricked early in their development to ensure a reasonably secure progression of economic growth reflected in a market price. If you are going to select capitalism as your system of choice then a socialised capitalism which does not allow proprietary advantage to insiders is surely preferable. Going back to our stockmarket. In the year 2000, our market is at 5000. Increasing at 9% per annum until 2010 would lead to a price of 11837. The psychopathic operations refuse to accept this structure. These people prefer a market that oscillates with controlled volatility resulting in repeated booms and busts that their models (which are able to utilise proprietary inputs which all other market traders and participants can only dream of possessing) are able to largely predict. Consequently, the line that was initially drawn between 5000 and 11837 may be significantly lengthened eg 5000-9500-6500-13000-9500-15000-10500-11837 for example. With extremely privileged inside information and knowledge, the core market participants are able to massively increase their profits at the expense of the amateurs. If the truly perfect model were to exist (which it doesn't, of course), the supreme trader would be able to yield a personal advance in price of 28837 as opposed to a socialised 6837. An 80% efficient model would still yield 23070.
Of course, it is all rather more complex than the simple picture that I'm describing but the template is valid. This is neither the time nor the place to be more expansive in my explanation.
All of the above are examples of factors that must be taken into account when approaching financial markets analytically from a chart/trend perspective. However, my point is that one can only truly rely on such analysis with a healthy input of Bayesian reasoning.
Got it?