Hi, I've started to grow interested in trading when I discovered about automated trading. I have a IT degree and know about programming, but started to get info about trading about 2 months ago.
I started programming some algorithms in Ninjatrader, discovered how to do a basic backtesting and there seem to be some potential, but I can't understand yet the whole picture.
Searching for info I found that backtesting is not necessarily reliable, on the contrary it's not obvious how to get a realistic picture of how an algorithm would perform live.
Another problem is to find a reliable broker that handles well the various problems that may arise (an order not taken, a spread widened to catch you on stop loss, etc...)
I also have trouble figuring out what all the costs are and if there is a way to make intraday automated trading profitable. Also, I can't figure out what are the results required to even consider an algorithm a good one. (Later I'll post some backtesting screenshots.)
I must say that my goal isn't to make 500€ day nor to become a millionaire. Since a year ago I started a daily job that I like, and I'm searching for a investment method that gives a decent revenue to add to the salary.
I know I'll have to study for months and to test even more, and that's where I'm heading to.
Let's start with backtesting and costs.
What should I be careful about when I test an algorithm?
What are the costs I'll have to face? I mean commissions, spread, margin and so on, but not connection or one-time fees. Everithing that goes in a backtest to make it realistic.
I would recommend you start by reading all the books you can on strategy development. That will answer many of your questions. Pardo, and Tomasini wrote very good books. Bandy has some good ones too.
The biggest thing to realize is that there are tons of mistakes you can make. I am speaking from experience here, because I've made a ton myself.
I would not jump into testing yet, because most software encourages bad practices.
I would not pay for any courses, etc just yet, except for Elite membership here. Down the road, you might find benefit in paying for quality material, but starting out there is plenty of free or nearly free material out there.
If you have any questions please send me a Private Message or use the futures.io "Ask Me Anything" thread
The following 5 users say Thank You to kevinkdog for this post:
Sure, I forgot to mention that I registered to this forum exactly because searching info in it I found that it has many topic covered, although most of the info I can find is indeed scattered around.
Obviously I wasn't hoping in a all inclusive answer or a guide, but just some clues about what info I should focus on and what are the viable routes...
I'll look at his posts, thank you. I imagine that it's way too soon to ask for precise suggestions to a professional, I need to get a basic knowledge to even formulate decent questions...
Thank you, I'll start from there and search around to see if I can get a picture on how to start moving to get it done the right way.
Thank you, I'll look for these books, and there is indeed so much free material I need to read first that I'll be busy for a good while before starting to need more.
The little testing I did just got me curious, but also made me understand that I need to learn a lot about how to backtest before considering the results I get realistic in any way.
Last edited by tndlux; January 22nd, 2016 at 02:36 PM.
The following 2 users say Thank You to tndlux for this post:
I started reading your topic and I must say that I lack the basic knowledge to understand it fully. I need to study the basics first.
I found on amazon the book "The Evaluation and Optimization of Trading Strategies" by Pardo, is it a good start?
Just for information, I'm attaching the backtest results I had for a (very simple) strategy I developed.
1 Minute bars, 1 year backtest on EURCHF forex, trading 1000 units per trade.
It seemed to me that it could be worth giving it a better look, but once I add the commissions and slippage it crashes and burns badly. I don't post these results because the account just goes to 0 in a flash.
The commissions I used are the standard for NT, 0,04$ per 1k lot, and to get a 1k lot I used 1000 as quantity on entering orders. Is it correct?
I'm asking just to know if I understood how it works correctly, it's a basic concept and I'd like to have it covered.
Hi guys, I just finished reading "The Evaluation and Optimization of Trading Strategies" by Pardo, I will have to read it again once I have a better understanding of the basics to completely put it in practice, but surely it gave me a very good idea about where to start when evaluating a strategy.
Honestly though I lack the basics on how to design a strategy. What I mean is that now I basically plot on a chart different indicators and try to see if i find any combination that seem to represent a good signal, or I create some indicators myself to see if they have some usefulness, but it's all based on logic and intuition, I don't feel to have a solid theoretic base to get my assumptions from.
How can I get an understanding on how to design strategies?
Read some books, watch some vids, ask questions.
Understand volatility and and how it relates to draw downs.
Select a time frame you want to trade (best systems are time frame independent ie same rules apply (mainly) in any time frame).
Develop a strategy that works for you.
Use it with real money for a year or 2 or 3 or 4 or etc.
Learn how what you are trading works.
Account for random spikes in the market.
Work out what you can do when it all goes wrong.
Try and automate that.
There is no quick path as there are 1001 ways to trade a market.