Neophytes in the art of trading systems development will often leap into the problem before developing a proper understanding of those tools they are using or many of the common problems associated with developing a trading system.  This article and many of the subsequent Education, Analysis, and Key Levels articles will cover ways to make sure your trading system is robust.

The Idea:

Your trading system should start with an idea – a reason you believe that the markets may be inefficient.  Many ideas already exist in the academic literature, and you can use or expand on those.  If you do enough reading and thinking, you are sure to also come up with some completely original ideas. 

It is generally inappropriate to start developing a trading system without having an idea to start with.  The alternative – data mining – is highly inefficient if the scope is too wide (you have absolutely no idea what to study), and interpretation of the results may be extremely difficult in many cases, since high statistical significance following data mining is likely to occur by chance alone (in sample).  As a result, data mining often uncovers spurious relationships.

To reduce the probability that your work is uncovering a historic anomaly and increase the probability that there is causation, you should start with an idea that has valid reasoning.

A simple example:

Company insiders have excellent access to public information on the likely success or failure of the company.  They also have a large stake in the success of the company and as such, have an incentive to interpret and understand the information that they have about the business.  In addition, insiders may have access to information that has not yet been distributed to the public.

While it is illegal for insiders to trade on private information, it is practically difficult for legal authorities to detect and prosecute based on such trading, so insiders may trade on it anyway.  Insiders are allowed to trade based on the public information which they probably process quite effectively. 

Since insiders are likely to have a better understanding than outsiders about company prospects, trading by insiders may reveal new information about the company.  Insider trading information may be less obvious of a factor to most investors than fundamental information about the company, which is where most analysts place their focus.  If in fact investors do not fully appreciate the value of the new information, abnormal returns may be generated by trading based on it.

An empirical investigation of insider trading has been performed in many academic studies.  Results suggest that abnormal returns may indeed have been generated by trading based on such information.

Analyses of HOLDRs

BBH:  Bearish.  (Biotechnology)

SwingTracker
MrSwings Real-Time Stock Charts RISK-FREE TRIAL featuring one-click access to Larry Swing's profit-generating indicators - Force Index, EquiVolume, True Strength Index

BDH:  Neutral.  (Broadband)

BHH:  Neutral.  (Business to Business)

EKH:  Neutral.  (European Stocks)

HHH:  Neutral.  (Internet)

IAH:  Neutral.  (Internet Architecture)

IIH:   Neutral.  (Internet Infrastructure)

OIH:  Neutral.  (Oil Services)

PPH:  Neutral.  (Pharmaceuticals)

RKH:  Neutral.  (Regional Banks)

RTH:  Neutral.  (Retail)

SMH:  Neutral.  (Semiconductors)

SWH:  Neutral.  (Software)

TTH:  Neutral.  (Telecommunications)

UTH:  Neutral.  (Utilities)

WMH:  Bearish.  (Wireless)