The week I started working and receiving a paycheck I began to ask questions about how I could invest what I was earning. I have always found a fascination in the idea that you can take money and have it accrue into additional dollars by making the right investment choices.
I began trading in the stock market when I was naive to say the least. Asking friends and family about investment ideas usually resulted terribly so I began to read about trends, sentiment, indicators all claiming to be the holy grail. Some years I made good profit but the anxiety and large swings of positions both in and out of profit kept me up at night. When positions turned or the holy grail indicators no longer worked I would deny reality and my strategy became that of hopes and prayers. It was a lesson I am glad I learned at a young age.. It has helped lead me down the path I am currently on and highlighted the necessities of risk management in trading.
After many years of the ups and downs I began looking into automated strategies. At first, I started playing with backtesting programs and modeling simple strategies based on the known and available indicators. I focused my efforts on large cap & small cap stocks. After modeling every holy grail system I could think of and trying countless approaches with the majority ending in failure in the long run I started to think outside of the box.
I began employment at a medium sized software company after college. It required me to live for a year over seas in the middle east and I had little to no access to the markets. It was here that I began developing an automated futures scalping strategy because I wanted a reliable system that I could be hands off if I was over seas that could generate additional income. Its been 4 years since I decided to take a shot at automated trading. The modeling process has been a lot of trial and error. The goals and expectations I have for the strategy are fairly high. A strategy that is profitable on all forms of futures (index, commodity, and currency), a strategy that day trades only, it does not require optimization of parameters and to never have a peak to valley draw down greater then 20% on the account.
I am modeling my strategies on three different types of futures portfolios each portfolio consists of a starting balance of $25,000. Portfolios are as follows: Index Futures, Commodity Futures, Currency Futures.
I am still testing my strategy and have not committed physical money to it. This journal is to track progress of my strategy as well as any future changes and the notes for those changes.
Currently the strategy is a mixture of trend following, mean reversion and light machine learning. Its foundation is built on multiple indicators that I have coded myself and it does not use the typical indicators for looking for entries/exits. It performs better in volatile markets, the most difficult part has been generating profit in flat markets. The entries and stops are generated dynamically and not based on a target. The strategy uses range, better renko, and one minute bars.
I have built the strategy with the goal to limit as much risk as possible and still be very profitable. It is designed with trading a single lot per future and does not chase or martingale for higher percentage wins. It can be opened up to adding additional lots in the future but currently with the model accounts having a $25,000 starting capital I will maintain single lot contracts and will not open it up to adding positions to winners or losers.
Feel free to ask any questions. We will see where this journey goes .
I will be posting profit/loss summaries daily and the creation of this journal is to compete in the FIO challenge for Sept 2016.
I will be VERY interested to see how your auto-trading thingy works after 8-12 months or so of data. The one-day results are impressive, but if the system can stand the test of time, I will be all over it. I like the stats!
I made some major changes to the auto-trader in the last few months and decided to drop the ball on the journal as there was a lot I wanted to do to it. I mainly wanted to take the underlying system and develop it to try and pass the 50k combine. I have put into place some strict rules to limit draw downs that match with what the combine offers.
-JTEconNews: Modified this to track an event area to exit positions (this is part of the FTP rules)
-rfkPositionMonitor: Modified this to track open profit loss and the number of positions that are open (this is for combine)
-Daily Loss limit of $1,000.
-Email alerts to open/closed positions and P/L of each position
I needed to track combined open/close profit loss and to limit the max positions open to 5 contracts per the rules of the combine.
I ran into some issues with the JTEconNews modifications but corrected them and am now seeing that it is functioning correctly. I am going to post daily results here to track the journey and be more diligent of this journal from now on. If I do not run into issues and the system is showing success after 30-60 days I will be looking to take part in the 50k combine.
Results of 3/20/17 - $+198.75
I am running a basis of $14 commissions round trip (this is high) but helps with slippage as well. Would be more profitable with around $4 (commissions + exchange fee's)
Last edited by PlayingGoFish; March 20th, 2017 at 06:52 PM.