I would like to explore this strategy using etf's.
Eugene offers the following narrative:
QUOTE:
Efficient Rotation is a non-Intraday strategy that continuously holds n stocks of a DataSet that have the highest Kaufman's Efficiency Ratio value averaged over three time frames. Although the number n is determined by the n Symbols Strategy Parameter, you must adjust Portfolio Simulation Mode sizing appropriately for the number that you select. For example, with 3 symbols you might set 33.3% of Equity sizing.
I used an etf data set of 8 etfs, which generally have low inter-correlation. EPP EWJ GLD IEV ILF IWM SPY TLO. I set n=1, equity position to 90%, and margin factor as high as 5 to 1. Data range begins 5-15-2006. With this range all symbols have the same number of bars in order that performance results are always the same no matter which symbol is highlighted to initiate the program. Periods are the same as default except for p3 which is 90 and days held is 45.
My assumption, overly simplistic and incorrect,is that the program ranks the data set by a Kaufman metric and that every n-days, if n symbol is only one, it buys whatever ticker is the highest. I can tell that my assumption is wrong because when running the program there are as many as five symbols that are open at the most current date.
Another point of confusion is that no matter how I adjust the equity percentage or the margin factor, I am still dropping trades.
I see considerable value in the basic structure of this coded program. I am wondering if Eugene or someone could offer additional descriptive narrative for someone that does not understand code, which may require considerable forgiveness.
Within the code I see the following:
QUOTE:
//Close positions that are not in the new lowest 3
for(int pos = ActivePositions.Count - 1; pos >= 0; pos--)
By any chance, should the descriptor "lowest" be "highest?" And, if the n symbol was only one would the numeric "3" be written as "1?"
Well, this is a start. Thanks.