Designing an Automated Trading Program

[expiated]

I already have a journal that I named “Numerical Price Prediction Challenge” and another titled “Let Your Profits Run!” But the purpose of the first was to alter my system to accommodate a 25-pip minimum profit requirement, and the goal of the second was to develop the self-discipline necessary to refuse to allow myself to exit trades until and unless I was stopped out or until my predetermined (and relatively distant) destination had been reached.

However, these approaches are not ideal strategies in my mind, given my understanding of price action. And since, in developing an automated system, I will be looking to convert only the preeminent aspects of my methodology into code that my computer will understand and run by looking for trades that strictly adhere to the corresponding rules and automatically placing orders with a broker, I am starting this new thread to keep track of my personal thoughts and anything else related to this endeavor.

 

[expiated]

General Guidelines:

 

[expiated]

My system will be specifically designed to trade foreign currency pairs, and one aspect of my methodology will reflect certain steps practiced by Nick McDonald…

Step 1:
Plot the 10-, 20-, 50- and 200-period moving averages on your chart.

Step 2:
Notice especially when the 10-, 20- and 50-day moving average lines are fanning out on a strongly trending financial instrument.



Step 3:
When the price candles pull back into the space between the 10-period moving average and the 20-period moving average, we are in the BUY ZONE. Note however that the trigger for taking the trade is a small candle in the buy zone. Once the small candle has formed, the trigger to enter the position is one pip ABOVE/BELOW the high/low of that previous small candle.

More specifically, the practice of identifying fanning moving averages will be a key feature in the designing of my algorithms.

However, I will not be trading based on daily charts, and the fanning pattern I will be looking to identify will involve painstakingly selected moving averages—not your standard 10-, 20-, or 50-period SMAs. Moreover, the trigger for taking a trade will be based on intraday price ranges and price action rather than candlestick size or whether price is one pip beyond the high or low of a given candle.

 

[Ayn Rand]

1. Determine a market you would like to trade. What is it about this market that you like?
2. Identify as many players in your chosen market as possible. It it helpful to be able to measure the influence or each group of market participants on the market. If you are doing this what measure of influence are you using?
3. Why would 10/20/50/200 MA be relevant for these participants? Pertains to trading horizon and reaction time to incoming stimulus.
4. Is relevance, static or dynamic with respect to some measure of time - day/week/month/quarter/prior? Does it change or is it influenced by ex post movement?

5. "When the price candles pull back into the space between the 10-period moving average and the 20-period moving average, we are in the BUY ZONE."

a. Is this a long only setup
b. Explain in detail the significance of the pull back into the documented space? When there is a pull back what is the financial implications for the market participants?

6. When would you expect your methodology to optimize (work best)? Under what conditions would your methodology underperform?

Take all of the above and apply to J. M. Keynes and the "beauty contest winner"?

 

[expiated]

5. "When the price candles pull back into the space between the 10-period moving average and the 20-period moving average, we are in the BUY ZONE."

a. Is this a long only setup

An oversight in copying and pasting. In my personal notes I changed it to TRADE ZONE.

 

[expiated]

Take all of the above and apply to J. M. Keynes and the "beauty contest winner"?

Andrew Allentuck commented that John Maynard Keynes’s analogy of likening investing in shares to a judge’s task of picking a beauty contest winner on the basis of who the greatest number would think is prettiest conveys the point that the stock market is a competition of companies seeking money, with the most appealing companies selling more shares and getting more money per share sold than other firms seen as less attractive. He states that over the long run and over a market cycle, the daily and weekly news, month-long trends, and even longer waves hardly matter.

However, given that I will be trading foreign currency pairs—not equities—and only on an intraday basis, I’m not sure how much all of this will actually apply.

 

[expiated]

2. Identify as many players in your chosen market as possible. It it helpful to be able to measure the influence or each group of market participants on the market. If you are doing this what measure of influence are you using?
3. Why would 10/20/50/200 MA be relevant for these participants? Pertains to trading horizon and reaction time to incoming stimulus.

My system is based on price action alone. It’s all about the numbers. To be honest, the “why” behind price behaving in a particular manner is not of great concern to me. Rather, I am looking to establish a high level of statistical probability. Given what price is doing now, how high are the odds it will do thus and so next based on an objective analysis of past performance under the same set of figures? Is there a quantifiable pattern, and what is the predictive value of the corresponding forecast model over time?

 

[expiated]

What is it about this market that you like?

I find the Forex market to be extremely predictable. Also, the amount of initial capital required to participate is extremely small compared to most other markets. And finally, the fact that the Forex market is dominated my multinational institutions who trade toward key levels where the supply of liquidity offers them a certain amount of protection, who are interested in where large groups of stop orders are sitting on an aggregate level, who must make trades in accordance with their order books, and who must balance their books on a regular basis, gives the market a certain orderliness that facilitates the success of observant traders with insight as to why the market ranges or breaks out at certain levels, and why key levels act as support or resistance.

 

[Cheesy]

My system is based on price action alone. It’s all about the numbers. To be honest, the “why” behind price behaving in a particular manner is not of great concern to me. Rather, I am looking to establish a high level of statistical probability. Given what price is doing now, how high are the odds it will do thus and so next based on an objective analysis of past performance under the same set of figures? Is there a quantifiable pattern, and what is the predictive value of the corresponding forecast model over time?

Have you looked at the Dynamic FIb Moving Average system? That fits your style, there are many variations of it and I have deconstructed the original which was invented by Andy Shearman I beleive, and then built my strategy around this idea as well, been using it for many many years now.

 

[Ayn Rand]

"I find the Forex market to be extremely predictable."

Do you know how many people on this forum find this to be true? 0

Pick any recent or even better future occurrence where you could predict what was going to happen and why.

Anyone able to do this would be King at least for A Day.

ALL IN