The research, of course, could end up following any number of paths. In the beginning, however, it will start by looking at the actual performance of retail trader accounts. This will help at least go a long way toward answering the question of what fraction of traders actually sees consistent profitable performance. This is about as empirical as it gets.
So we have trader “A” who in the data series that you examine, is a winner [or loser] and is selected as a representative of the data set. The immediate problem is that trader “A” might fall into a different data set in a different historical period.
The empirical data that you are examining are human individuals that do not hold constant properties as do ‘things’ in the physical sciences: water is always 2 atoms of hydrogen combined with one atom of oxygen, and under variable conditions, pressure, temperature, etc. provides quantifiable and consistent results. The same is categorically not true of our trader “A”. Thus even this ‘limited’ empirical investigation is fatally flawed.
When we move onto the ‘Behavioral Finance’ aspect of the research, the flaws start to grow exponentially.
Where BF comes in at this point is trying to explain individual performance, with the theories to-date mainly focused on why individuals lose in the markets. Over-confidence is a favorite, though I’m not a big fan of that one. There are a number of other “biases”, however, which can be contributory. I believe that is what you would call the “causitive” aspect of the work. Note, however, I am not talking about markets at all here, though there is the opportunity to see if/how market conditions can impact individual trader performance.
The answer to your research question can be answered in three sentences: individuals have value scales which drive required ‘ends’. Individuals employ means to achieve those ends. Not all the means employed are appropriate, and therefore do not achieve the ends.
Essentially, your research project is seeking to investigate the ‘means’ employed by individuals. As we have already seen, individuals learn and progress, they also retrogress. In addition, because the ‘market’ is a process made up of thousands, millions of individuals, the market is no more a constant than are the individuals. The data from one historical market are completely irrelevant to a future market.
Assuming I can demonstrate that some fraction of the population demonstrates non-random profitability, the door is then open to compare and contrast them with the rest of the sample and see if there are statistically meaningful variations in the measurable variables.
Which rather contradicts your previous assertion, and proves mine. I haven’t yet, but I will address the entire ‘Behavioral Finance’ area.
Week 14 was once again profitable. The intra-week dip could have profitably been bought, and the buy & hold strategy continued to increase the open profits.
From the first part of the presentation, where I looked at the type of market, we can now move into a brief discussion as to technical indicators. Technical indicators come in two basic categories: [i] trend following and [ii] stochastic.
Trend following are of course, in a trending market, the indicator that you want to weight the heaviest in your analysis. It should go without saying then that identifying correctly the type of market prior to the selection of a technical indicator should be the first priority. As stated, in a trending market, choose a trend based technical indicator.
In a range-bound market, the stochastic based indicator is the better choice. Again, the priority is first correctly identifying the type of market that you are actually trading.
For the more mathematically inclined, simply consider the calculations that underpin the various technical indicators and you will immediately be able to identify the differences between the two types of technical indicators that use historical prices as their data set.
Essentially all technical indicators that use the time series of price, or volume data, generated within the market, will fall into one of the two basic distinctions; trend following or stochastic.
I’m going to post last week’s newsletter in a few minutes. I’ll have the results posted here after the market close, so far, looking good, unless we have a flash-crash in the last hour of trading.
Again, just to remind you the underlying thesis:
In fact, that will be the focus of my dissertation. My research proposal for admission was on the subject of individual retail forex trader profitability and performance. It fits into the general area of Behavioral Finance, which is basically the counter-point field to classic efficient markets theory.
I’m really looking forward to delving into the data (I’ll be using a big set of trader transactional and performance data for my studies) and the sort of results that come from it. As much as this will be academically oriented work, I see lots of opportunity for it to be applied in the practical arena.
Delving into the data and ‘Behavioral Finance’ are contradictory, and doomed to failure. First, let me look at the ‘data’ component of the thesis.
There is a confusion between the relationships of certain events. This is particularly true when we talk about the ‘market’ which, of course the Fx market is, a market for the commodity of fiat money. Some schools of thought hold that markets conform to well established empirical laws, when in reality, there is no such thing. Market events conform only to necessary and logical a priori axioms, or economic laws.
Empiricism represents, in the study of markets, the fundamental confusion between the categorical difference between theory & history.
Empiricism is defined in the following way: knowledge regarding reality, must be verifiable, or according to Popper, at least falsifiable by or through observational and experimental experience.
Observational knowledge, and we are referring to observational knowledge of a market, can only lead to contingent knowledge. This is due to the true statement that the knowledge could have been different, without any necessary change in the circumstances or conditions. We know this is true because we are dealing with individual humans, who, have values that drive action towards fulfilling ends via the employment of means. These values are not fixed, they fluctuate from second to second.
Empiricism of the physical sciences, Physics, Chemistry, etc. allows for, based upon constant observational and experimental experience, or data, to make causality statements: If “A” then “B”, in addition they allow quantitative measurement, which further allows, if more “A” then more [less] “B”.
It is the nature of empiricism that the relationship between history and theory becomes entwined, confused and conflated, so much so that in point of fact they become the same thing. The only difference is that the empirical record, or the historical record, refers to history, viz. what actually happened, and empirical theory refers to prediction of the future, through a projection of the empirical history.
As already stated, in markets, where the empirical data, or market history is contingent and not necessary, that empirical theory, based upon empirical data, is invalid. It is a nonsense on stilts.
I understand from the brief background, provided by the post lifted from RhodyTrader’s blog, that the PhD will utilize empirical data from 1 or more data bases. The PhD will further examine Fx trader means and ends through a ‘Behavioral Finance’ lens. The game of data analysis, the empirical [historical] examination of the data requires that you look for events or actions, or both, that you consider causative. The number of variables can be few or many. Usually today, using a computer, you statistically crunch the numbers to examine functional relationships between the causative hypothesis, the events, the actions and any other identified variables in a linear, curvilinear, recursive, non-recursive, additive, multiplicative relations etc.
I guarantee that relationships will be found. They will unfortunately, for empirical theory, be worthless and simply a nonsense.
Well as I have been alluding to in my newsletter and on the blog, the market, in the absence of viable alternative investments, and under the continuous money & credit creation that has been termed QE, QE Lite etc, the market has only pretty much one direction.
For those who don’t follow the Federal Reserve releases & policies, and who simply rely on technicals or other, the message has been the same, barring some very minor fluctuations in the trend intra-day and/or over a couple of days this week, the trend has been slowly, but relentlessly higher.
1) 2.2 percent is the average interest rate on the U.S. Treasury’s marketable and non-marketable debt (February data).
2) 62.8 months is the average maturity of the Treasury’s marketable debt (fourth quarter 2011).
3) $454 billion is the interest expense on publicly held debt in fiscal 2011, which ended Sept. 30.
4) $5.9 trillion is the amount of debt coming due in the next five years.
For the moment, Nos. 1 and 2 are helping No. 3 and creating a big problem for No. 4. Unless Treasury does something about No. 2, Nos. 1 and 3 will become liabilities while No. 4 has the potential to provoke a crisis.
In plain English, the Treasury’s reliance on short-term financing serves a dual purpose, neither of which is beneficial in the long run. First, it helps conceal the depth of the nation’s structural imbalances: the difference between what it spends and what it collects in taxes. Second, it puts the U.S. in the precarious position of having to roll over 71 percent of its privately held marketable debt in the next five years — probably at higher interest rates.
First Among Equals
And that’s a problem. The U.S. is more dependent on short- term funding than many of Europe’s highly indebted countries, including Greece, Spain and Portugal, according to Lawrence Goodman, president of the Center for Financial Stability, a non- partisan New York think tank focusing on financial markets.
The U.S. may have had a lot more debt in relation to the size of its economy following World War II, but the structure was much more favorable, with 41 percent maturing in less than five years, 31 percent in five-to-10 years and 21 percent in 10 years or more, according to CFS data. Today, only 10 percent of the public debt matures outside of a decade.
Based on the current structure, a one percentage-point increase in the average interest rate will add $88 billion to the Treasury’s interest payments this year alone, Goodman says. If market interest rates were to return to more normal levels, well, you do the math.
Some economists have cited the Treasury’s ability to borrow all it wants at 2 percent as an argument for more fiscal stimulus. Why not, as long as it’s cheap?
Goodman says the size of the deficit (8.2 percent of gross domestic product) or the debt (67.7 percent of GDP) is only part of the problem. The bigger threat is rollover risk: “the same thing that got countries from Portugal to Argentina to Greece into trouble,” he says. “It’s the repayment of principal that often provides the catalyst for a market event or a crisis.”
The U.S. is unlikely to go from all-you-want-at-2-percent to basket-case overnight. That said, policy makers would be wise to view recent market volatility as a taste of things to come.
Talking to Goodman, I was reminded of the Treasury’s standard sales pitch before quarterly refunding operations during periods of rising yields. Some undersecretary for domestic finance would be dispatched to tell us that Treasury expected to have no trouble selling its debt.
I had an equally standard response: At what price?
That seems particularly relevant today. The Federal Reserve purchased 61 percent of the net Treasury issuance last year, according to the bank’s quarterly flow-of-funds report. That’s masking the decline in demand from everyone else, including banks, mutual funds, corporations and individuals, Goodman says.
Of course, Fed Chairman Ben Bernanke might look at the same numbers and see them as a sign of success. His stated goal in buying bonds is to lower Treasury yields and push investors into riskier assets.
Free to Borrow
Then there’s the distortion in the relative value of stocks versus bonds to worry about. Using the 10-year cyclically adjusted price-earnings ratio and the inverse of the 10-year Treasury yield, Goodman says the relationship hasn’t been this out of whack since 1962.
The Treasury isn’t unaware of the rollover risk. At the same time, it’s trying to accommodate the increased demand for “high-quality liquid assets,” such as Treasury bills, as required under new international capital-and-liquidity standards, says Lou Crandall, the chief economist at Wrightson ICAP in Jersey City, New Jersey.
In fact, when Treasury bills carry a negative yield — when investors are paying the government to hold their money for three, six or 12 months — borrowing “more is better,” Crandall says.
Still, the dangers are very real and were highlighted by Bernanke himself last week in the second of four lectures to students at George Washington University. Explaining why the decline in house prices had a greater impact than the drop in equity prices less than a decade earlier, Bernanke talked about “vulnerabilities” in the financial system. Too much debt was one; a reliance on short-term funding was another.
I doubt he had the Treasury in mind when he was explaining how the subprime debacle morphed into a global financial crisis, but the U.S. government would be wise to heed his advice. Currently its demand on the credit markets for annual interest and principal payments is equivalent to 25 percent of GDP, Goodman says, 10 percentage points higher than the norm. That’s real money. And with the federal budget deficit projected to top $1 trillion for the fourth year running, the funding pressure is bound to increase.
So the next time you hear someone say the Treasury can borrow all it wants at 2 percent, tell him, that’s true — until it can’t.