Search results

It is a good point. After reading all the comments (the backtesting point was the most discussed), I can't disagree that there is value in creating your own backtesting engine, what I meant is - watch out for becoming a backtesting engine developer in the process. For guys with software dev...

Rob, you should publish your list. I'm sure people will appreciate it. #2 is the most controversial it seems. I did write my own too and there was a value in it, but also a good deal of distraction. Perhaps I should combine it with "80% spent on your strategy development, 10% on experiments...

Don't quit your job (same as advice given by a Market Wizard of systematic trading - @mhparker) Don't write your backtesting engine Expect to spend 3-5 years coming up with remotely consistent/profitable method. That's assuming you put 20h+/week in it. 80% spent on your strategy development, 10%...

Or rather unlearn a concept of risk management and just start betting a farm on stocks and bitcoin only going up. One guy I know told me last week that he feels bad about making only 7% a week prior. He started trading 4 months ago. Another one told me that after careful analysis of his...

As you can imagine, many people make those claims. Perhaps, a mentor, actively selling his services will reach out as that would be part of their routine job, to talk to potential clients. Otherwise, not likely, as typically, this is a waste of time. It will help if you have a series of posts...

Great strategy! Congratulations.

If you wanna be accountable - post your findings after the weekend! :) When I migrated all my code from AB to RealTest it became so easy/safe to test ideas that I even coded couple in a response to somebody asking a question on ET. Which I would never do with AB, too time consuming and error...

@Nobert - thank you for posting. Such an unusual mix for this forum. Love the art, love the music selection. Would follow this thread for the music picks alone. Val

I have Pingdom checking my bot's health API every minute and SMS/email if returns wrong code or can't reach. There are many services like that. Some for free. Val

Generally speaking, to get filled anywhere close to high/low you would have to be well under 1% of trading bar's volume. Wouldn't hurt to include a bit of slippage in your assumptions. Then try to trade live with smaller size than you could at first, capture your fill stats and increase size...

@yc47ib, realistically, if you are in this long term, you will choose not based on how easy it is to get started, but the core feature or two you need. Sounds like you decided on Alpaca because of API+commission free. IB doesn't compete there, so just give Alpaca a try and see if strategy...

Since I don't know your background I will answer in a bit general sense. I've read hundreds and there is no single one where I would take a strategy from and start trading it. At some point I just started feeling that I reached the end of the universe when it comes to yet another strategy or a...

In "Fooled by Randomness", Taleb portrays a happily retired dentist who expects to earn 15% annually on his investments with a 10% error rate (or volatility) per annum. This "translates into a 93% probability of making money in any given year. But seen at a narrow time scale, this translates...

You're living many traders' dream. Markets keep paying you to continue your study of them :D

Find 1000 trades like that. Run a simulation for each of your targets. Pick whichever combines equity you like the most.

JVM is the runtime. Kotlin compiles into same bytecode as Java and runs on JVM. No difference for what you are asking. Nothing is without failure. Everyone, without exception, faces a possibility of a loss year unless they don't trade at all. 2 orders of magnitude means you are trying to...

There will be few practical limitations that will make you application side latency not matter as much as you think. Likely limitations of your data feeds, your network latency and lack of direct access to exchanges. In total they will be 2-3 orders of magnitude more than what you will save by...

public static void main() { // your code here } Good news - performance is likely going to the last problem you need to be worried about when it comes to retail level algorithmic trading.

I splurged on coding this weekend. To save a bit of time on doing weekly/monthly routine I now automated PL/DD graphs generation. In the past, once a week I'd export trades from execution software, do some data massaging and put it into RealTest format, import new market data and run a test to...

A lot of trades with a positive expectancy is the best if you aim for returns consistency. But I can see how it can be tiring if you're doing all of that manually in front of the screen whole day. Maybe try to automate parts of it? Starting something like only exit, or some entries. My own...