Today I'd like to use my blog... well like a blog again, at least a little. As you might have noted, at the beginning of the year I stated I´d like to do more blog posts and use my homepage not for signals only. Well, as it sometimes happens: Knock, knock!- Who´s there?- Reality!
During the course of the year I was again pretty busy with my dayjob, my family, and way too little freetime in my evenings. At the same time I pushed the development of my code, as I had ideas over ideas that needed implementation and testing. So unfortunately I had the choice of doing teasing blogposts (plus the usual signal and weekly ichimoku posts) or try to do something productive. I chose the latter.
During my vacation I had plenty of time to think about the direction I would like to go, both with my system and my blog. Or if I even would like to go in any direction- closing the whole chapter also was an option. I decided that I went down this road way to far to just give up, and that I would lose more than a pleasing little hobby. So I tried to come up with a kind of a roadmap, parts of that I would like to share here.
The last release of my system was in April this year. Since then I have been really busy adding more and more features, to a point where I think the trading system part now is "ready". Up to this point, however, the software had no concept of "money". It doesn´t care if it is fed with Indices, Futures, Stocks, F/X, equities or whatever- it works agnostic of the underlying market. Which of course makes it nearly impossible to do a) proper position sizing considering the whole portfolio, and b) measure the performance over a large number of trades. The latter was done mostly percent-wise, which can give you some rough idea about profitability, but no "hard facts" in terms of money won/lost.
Since Summer I am now working on incorporation the concept of money into the trading system. Which is much harder than I thought, because I did not consider this problem when I started to write this system. But step by step I am getting closer to a state where the system is configured to have a so-called "account currency" and each market has to provide some information about its underlying currency, ticks/pips and a couple of other parameters. When done, the backtests will be much more informative and the live trading much more usable, as I will get not only the entry/exit signals and stop losses, but also the optimal position size considerung the currenct account balance and existing other positions.
I hope to finish this work in October, test it and write the necessary documentation and do a new release so everyone who is interested can play around with it.
The next logical step would be to revive the depots: The ones who follow my blog since it´s inception in April '16 might remember that in the original (spreadsheet/ macro) version of my system I did not only post the weekly analysis, but also had a kind of virtual deposit accout, where I tracked each signal and it´s outcome- basically I did papertrading. This is something I would like to do again, maybe even in different combinations (like one Indices, one for commodities, one for Forex, one mixed and so on). For this I have to automate the system a bit more, to a point where I basically won´t do anything at all: on my server a cron job will periodically update the data, run the system and spit out the results via email to me- I just have to execute what my system dictates.
For all that to work will propably take time until the end of the year. If everything runs as envisioned, I will start live trading with my own money. From there on, it gets more fuzzy ("Difficult to see. Always in motion is the future..."- Yoda). Maybe I use my system for social trading, maybe not. On the software side I have some big architectural changes in mind, which won´t bring new features, but allow greater abstraction and flexibility for different style trading systems. After all, my level of ambition is not to create a single Ichimoku-based trading system, but a framework to create any system I like. The Ichimoku one being the "test case" for the framework. But first things first, this is waaaay into the future.
I hope you liked the more bloggish blog post, sorry if you missed your daily ichimoku signal here :)
Wish you relaxed trading!
Comments powered by CComment