Sunday, February 22, 2009

Basing?

DJ plays out the 1-2-3 selldown pattern I suspected since Nov 2008.

Nothing is certain in the markets. I wont rule out a reversal to the upside, as UJ is shaping into a reversal setup.

Updated March:
UJ breakout and hit TP as setup, but DJ broke down instead. Did not "expect" the divergence. I am arguing with the price, obviously.

THE LESSON:
Divergence is there for a reason, respect it and not go against it.

Updated May:
That time was still in short mode according to system z. The yellow area resistance is indeed broken, but 1 month later. Incidentally system flashes buy signal on breaking this resistance, which sees DJ rallying from 7500 to 8500 till date.

Tuesday, February 17, 2009

Exits

Finished the following revisions:
Prototype 1.1 - entry strategy and basic exit
Prototype 1.2 - finetune system parameters
Prototype 1.3 - port system to shorter term

Current Agenda
1. test exit strategies below:
newlows, guarding, safezone, chandelier, t-indicators
2. compare method/results with other trading systems

March Agenda
1. research strategies to sidestep ranging markets
2. code in capital management module (2/6 rule)
3. write technical paper and plan future directions

Tuesday, February 10, 2009

War

Reading something on war. Thinking could this worst financial crisis in 70 years draw destructive parallels to World War III?

"Motivations for war may be different for those ordering the war than for those undertaking the war. For a state to prosecute a war it must have the support of its leadership, its military forces, and the population.

For example, in the Third Punic War, Rome's leaders may have wished to make war with Carthage for the purpose of eliminating a resurgent rival, while the individual soldiers may have been motivated by a wish to make money.

Since many people are involved, a war may acquire a life of its own from the confluence of many different motivations."

If you replace war with trend what can you arrive at?

...

Makes sense.. Confluence of personal agendas of traders / investors. Confluence of moving averages. Confluence of resistance / support lines, etc. Confluence of technical indicators. Confluence of fundamental economic conditions.

I reckon that's what make price tick in a trend.
Confluence of many different motivations.

Monday, February 9, 2009

Prototype 1.1

SGX buy/sell signals from system prototype.

Sunday, February 8, 2009

UJ Long

A trade using principle of price acceptance. CNA post.

NOBLE

A trade using principle of price consensus. As usual, post to CNA.


Updated May:
System already long at 100 in dec, still holding at 170 in may.

Saturday, February 7, 2009

Risk Mgm

A discussion on risk management.

Trend-Following
Imagine each unit of market price movement as an impulse - long impulse or short impulse. When you buy on a long impulse and there is a trend, another long impulse will appear and you raise stop accordingly (trailing stop). If there is no trend a short impulse will appear and hit your stop, you sell. In a few big trends you make huge profits, to cover small frequent losses and more, ie trend-following.

Gambler's Fallacy
At the casino, after a string of heads is tossed, what is the odds of a tail the next toss? Common sense let us think there is high chance of a tail because so many heads in a row is rare. That's not true and can be proven mathematically. The next toss is still 50/50 for a fair coin. The analogous context in stocks is if a stock keep dropping a lot, it does not mean its time for a rally.

Sunken Cost Effect
When a trader is at net loss, he will often take more risks than he would if he is at net gain. Because he has lost that much to market, he expect to make back that much from market. However, such mentality induces high-risk actions which backfire his original intention. The wrong mindset will make him continue to lose more.

Prospect Theory
Human are likely to take riskier actions in losing positions than in winning positions. Amateur traders like to cut profits short and let losses run. They are "scared" to let profits run away but "hope" that losses can become gains, contrary of probability. Recognise these emotions when it happens to you. Act on logic, not on emotions.

Babe Ruth Effect
The classic case for portfolio managers. Everyone in the team had net many winners and few losers but all made a net loss. Except one guy who has a small number of winners while most are losers. He made a net profit instead. How come? He cut losses short and let profits run. To survive longterm, keep taking lots of small losses.

Kelly Formula
Define the optimal risk ratio to maximize gains in shortest time. Kelly Ratio = win/loss ratio - [(1 - win/loss ratio)/(average gain/average loss)]. For eg, you can use 1/4 of Kelly Ratio to take a slower approach and hence lower drawdown. This is important. "Optimal" ratios can create big drawdowns which is psychologically unbearable to last until turnaround.

Two-Six Rule
This method from Alexander Elder protects a trader in two ways. Not risking more than 2% of trading capital on any single trade protects a trader from a big single loss (shark bite). Not risking more than 6% of trading capital in one month protects a trader from a series of small losses that hurts (piranha bites). If a trader lost more than 6%, his capital will stay outside market for rest of month.