31.07.2021 Views

Ultimate Algorithmic Trading System

Using automated systems for trading in stock markets

Using automated systems for trading in stock markets

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

The correlation coefficients range from +1to−1. A coefficient of +1 represents

a 100 percent correlation and a coefficient of −1 also represents a 100 percent

correlation, but in the opposite direction. The AD (Australian dollar) is highly

correlated to GC (gold), but not very much to the CU (Euro currency), and is highly

anticorrelated to DX (dollar index). If you were picking two currencies, then you

would probably want to go with the AD and CU. The BP is a little too noncorrelated

for my taste. So let’s pick the AD and CU to represent the currencies. Moving

onto energies the crude is highly correlated to both the heating oil and natural

gas. Unleaded is a different story; there is very little correlation there. CL and RB

(unleaded) will take care of the energies sector. The grains consist of soybeans, corn,

Kansas City wheat, Chicago wheat, corn, and a few other low-volume components.

According to the correlation matrix, a good mix would be the corn and KC wheat.

The financials are up next and are represented by 30-year Treasury bonds, 10-year

notes, 5-year notes, 2-year notes, and Eurodollars. All these markets are highly

correlated to each other, so there is no advantage/diversification by trading one

with another. Trading a Treasury bond with Treasury note is like trading two bonds.

Bonds are very shock sensitive to news events such as Fed and employment situation

reports. So the less volatile but still with some bang is the Treasury notes. The safest

soft commodity is sugar (SB). It trends well and seems less sensitive to crop reports

than coffee (KC) or cotton (CT). There is very little correlation among these three

markets so any one of the three would probably suffice. All of the metals in the

current portfolio, silver (SV), gold (GC), copper (HG), and platinum (PL), are,

as you would assume, highly correlated, so like the financials only one should be

chosen for this initial portfolio. Copper is the only nonprecious metal in the list and

used extensively in manufacturing so let’s pick it. The initial portfolio will consist of

the following markets: AD, CU, CL, RB, TY, C, KW, SB, and HG. You will notice

that the stock indices were left out. The indices are a totally different animal. In an

interview with Richard Dennis by Arthur Collins, the King Turtle stated:

With the exception of the S&Ps, they’re all the same. You have to

treat them the same, for no reason other than sample size. You can

optimize each system in each market individually or you can optimize

them together. It’s better to optimize them together.

—Technical Analysis of Stocks and Commodities, April 2005

Issue

101

COMPLETE TRADING ALGORITHMS

An Act of Futility?

Okay, we now have a portfolio but on which algorithm should we risk our

capital? Also should we use the best-optimized parameters for each algorithm that

was derived from the global optimizations? Any of the four algorithms could be

used—they all showed positive expectancy. The ‘‘optimal’’ parameters that we

www.rasabourse.com

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!