“I'm currently enrolling myself to do a PhD in Behavioural Finance. I have a number of research topics that I have in mind within this discipline.

“Based on my personal trading experiences, I have learned a lot about myself in terms of discipline, emotion, and herding.

“It was mentioned in comment last Tuesday that market histories are best viewed using a chart (I agree) and that markets undergo cycles where trading errors occur.

“Behavioural research over a time series of, say, 30+ years will show up many cycles. Thus, the problem I'm encountering is testing a hypothesis that may be accepted over one sub-period or cycle but rejected over the entire sample period. I do not want to take a traditional academic and theoretical approach where I may be limited to testing linear models.

“Are there any areas in the field of behavioural finance that would be interesting and also helpful to trading? My own personal interests stems from fractal markets and fat tails.”

Eoin Treacy's view Thank you for this interesting email which I, suspect will elicit a response from the Collective. Your interest in the fractal nature of markets is very much in line with our own. The same patterns are as observable on an intraday chart as are evident over a much longer timeframe.

One of the reasons we distinguish between secular and cyclical bull and bear markets is because of the difficulty you speak of. For example, the S&P 500 has been in a secular process of valuation contraction and rising dividend yields for more than a decade. However, within that secular bear, there have been a number of cyclical bulls. Therefore I see no contradiction between a theory working over one timeframe and not over another, provided one acquaints oneself with the big picture.

At The Chart Seminar, one of our topics is the mistakes people make with charts. Number 1 on the list is myopia. There is a temptation, particularly when using leverage, to limit our examination to the most recent data. This leaves us exposed to a high probability medium-term event which one would not observe when concentrating on very short-term data. We advise starting with as much data as possible. When one talks about major bull and bear markets 30 years might not be long enough. For example, US Treasuries have been in a secular bull market for more than that period of time. This 90-year chart of the Dow/Gold ratio helps to illustrate three major bull and bear markets in stocks relative to gold. .

One of the primary characteristics of a crowd is that it thrives on contraction. If we look at any major crash, there will always be some patently absurd belief which was accepted as fact until just before the collapse. For the Nasdaq in the 1990s it was "earnings don't matter". In the credit boom it was "everyone can hedge against default". In the credit markets it is “we can fix a debt problem with more debt".

The acceptance of contraction helps define the loss of critical thinking though associated with group formation. When belief deteriorates, it is often a signal that the imbalance between supply and demand is reversing. The dissipation of previously large enduring crowds often results in the fat tail events you speak of.