Investment Madness

We are all prone to having psychological preconceptions or biases that make us behave in certain ways.  These biases influence how we assimilate the information we come in contact with on a daily basis.  They also have an impact on how we utilize that information to make decisions.

Our very own psychological biases have an impact on our investment decisions and affect our attempts at building wealth.

Psychological Bias	Effect on Investment Behavior	Consequence
Overconfidence	Trade too much.  Take too much risk and fail to diversify	Pay too much in commissions and taxes.  Susceptible to big losses
Attachment	Become emotionally attached to a security and see it through rose-colored glasses	Susceptible to big losses
Endowment	Want to keep the securities received	Not achieving a match between your investment goals and your investments
Status Quo	Hold back on changing your portfolio	Failure to adjust asset allocation and begin contributing to retirement plan
Seeking Pride	Sell winners too soon	Lower return and higher taxes
Avoiding Regret	Hold losers too long	Lower return and higher taxes
House Money	Take too much risk after winning	Susceptible to big losses
Snake Bit	Take too little risk after losing	Lose chance for higher return in the long term
Get Even	Take too much risk trying to get break even	Susceptible to big losses
Social Validation	Feel that it must be good if others are investing in the security	Participate in price bubble which ultimately causes you to buy high and sell low
Mental Accounting	Fail to diversify	Not receiving the highest return possible for the level of risk taken
Cognitive Dissonance	Ignore information that conflicts with prior beliefs and decisions	Reduces your ability to evaluate and monitor your investment choices
Representativeness	Think things that seem similar must be alike.  So a good company must be a good investment	Purchase overpriced stocks
Familiarity	Think companies that you know seem better and safer	Failure to diversify and put too much faith in the company in which you work

If you want to read more regarding human psychology and how it affects our trading and investment, please read the book “Investment Madness How Psychology Affects Your Investing and What to Do About It” by JOHN R NOFSINGER.

Smoothing The Bollinger %b

Bollinger bands were developed by John Bollinger and it consists of 3 lines derived from the moving average and standard deviation of closing price.

Sylvain Vervoort, a Belgium based retired electronics engineer who has been using technical analysis for more than 30 years, came out with an indicator SVE_BB that takes the Bollinger band and converts it to an oscillator that range between 0 and 100 (with overshoots).

The Metastock formula for SVE_BB is

period:= Input("%b period:",1,100,18);

TeAv:= Input("Tema average:",1,30,8);

haOpen:= (Ref((O+H+L+C)/4,-1) + PREV)/2;

haC:= ((O+H+L+C)/4+haOpen+Max(H,haOpen)+Min(L,haOpen))/4;

TMA1:= Tema(haC,TeAv);

TMA2:= Tema(TMA1,TeAv);

Diff:= TMA1-TMA2;

ZLHA:= TMA1+Diff;

percb:= (Tema(ZLHA,TeAv)+2*Stdev(Tema(ZLHA,TeAv),period)-Mov(Tema(ZLHA,TeAv),period,WEIGHTED))/(4*Stdev(Tema(ZLHA,TeAv),period))*100;

percb;

This indicator is often a leading indicator with smooth and clear turning points.  Bullish and bearish divergent signal is helpful in identifying entry and exit points. 

Let’s take a look when this indicator is applied to Sembcorp Marine.

Sembcorp Marine broke out of the Bollinger band in early October and it has managed to blow past the $5 mark to reach a high of  $5.20 on 10 Nov.   The stock has pulled back slightly and it looks set to test the 20 day moving average which is at $4.70 level.

The SVE_BB indicator did not make a new high together with the up swing.  This is a warning signal.  If this divergence coincides with a break of support level, it will be a strong indication of further down moves. 

The stock has to stay above the 20 day moving average.  If it breaks this line, the next support line is at $4.50.  

Two-bar Breakout System (2BB)

The 2 bar breakout system is designed to capture short term trends by going long or short depending on the pattern formed by the 2 most recent price bars.  The essence of this method is to look for reversal of the immediate directional momentum.

The rules of this method are:

Long trades

Enter:

1. Today’s low is less than the previous day low
2. Today’s high is less than the previous day high
3. Today’s close is less than the open
4. Go long at the next trading bar if the price goes above today’s high + 1 tick

Exit:

1. Use a trailing stop at previous day’s low

  

Short trades

Enter:

1. Today’s low is greater than the previous day low
2. Today’s high is greater than the previous day high
3. Today’s close is greater than the open
4. Go short at the next trading bar if the price goes below today’s low - 1 tick

Exit:

1. Use a trailing stop at previous day’s high

To see this method in action, let’s try it out on EURUSD currency pair. 

For long trades, the trading record is as such:

Position	Entry Date	Entry Price	Exit Date	Exit Price	Bars Held	Profit/Loss (pips)
Long	5/7/2009	1.3374	5/11/2009	1.3615	3	241
Long	5/18/2009	1.3486	5/25/2009	1.4016	6	530
Long	6/23/2009	1.3936	6/29/2009	1.4044	5	108
Long	7/9/2009	1.3938	7/13/2009	1.3941	3	3
Long	8/18/2009	1.4171	8/24/2009	1.4329	5	158
Long	8/27/2009	1.4349	8/31/2009	1.4281	3	-68
Long	9/30/2009	1.4646	10/1/2009	1.4574	1	-72
Long	10/8/2009	1.4737	10/9/2009	1.4703	1	-34
Long	10/29/2009	1.4841	11/1/2009	1.471	2	-131
Long	11/2/2009	1.4742	11/3/2009	1.4702	1	-40
Long	11/4/2009	1.4811	11/9/2009	1.486	4	49
Long	12/23/2009	1.4332	12/28/2009	1.4356	3	24
Long	12/31/2009	1.4361	1/4/2010	1.4315	2	-46
Long	1/13/2010	1.4546	1/14/2010	1.4454	1	-92
Long	1/22/2010	1.4144	1/25/2010	1.4147	2	3
Long	2/26/2010	1.3571	3/1/2010	1.3614	2	43
Long	3/10/2010	1.3636	3/15/2010	1.3758	4	122
Long	3/16/2010	1.3774	3/18/2010	1.3724	2	-50
Long	3/26/2010	1.3388	3/29/2010	1.3434	2	46
Long	5/27/2010	1.2345	5/30/2010	1.2267	2	-78
Long	6/7/2010	1.1966	6/17/2010	1.2252	9	286
Long	6/30/2010	1.2291	7/5/2010	1.2542	4	251
Long	7/6/2010	1.2558	7/9/2010	1.2623	3	65
Long	7/13/2010	1.2646	7/19/2010	1.2888	5	242
Long	7/22/2010	1.2914	8/4/2010	1.3145	11	231
Long	10/5/2010	1.3785	10/8/2010	1.3855	3	70
Long	10/28/2010	1.3879	11/1/2010	1.3955	3	76
				Total		1937

The method generated 27 long signals with 18 profitable trades giving a total return of 1937 pips.