I occasionaly check in here. Most guys are discretionary dudes talking about bullshit mindset rules, this guys approach comes close to how real quants handle their business
Spent a significant part of my life programming and a decent portion of that was automation in one form or another, started programming trading algos about 3 months ago and going strong rn, can't wait until its "perfected" 💪
When a trader has legit track records, it is definitely worth to listen to. Rather than wasting time listening to someone but has no track records, smells like a marketer than a trader
This was encouraging to see. I’m currently in the process of building my portfolio and plan to be using them within the next few months. It was helpful to know that there are others out there building their basket of EAs in a similar fashion. Great video, thanks.
Nice interview. I got funded at Darwinex a few years back, it took a rather long time. Perhaps because I didn't understand their home made criteria well enough, they're non-standard. Their broker platform is very good with low spreads. The profit split is very low compared to FTMO etc. As for multiple strategies, another option is to only use one low risk algo at a time on one asset. Then the correlation problem goes away. I've looked for such controling software with no luck. When it comes to predicting the market with AI, the big hedge funds have had zero success even though they employ dozens of PhD data scientists at a time. The problem I guess is over-fitting, you can fit historical data very well with all those parameters, but as the market is time dependent and never repeats the prediction power is low.
Great interview, as always! In case of a follow up, I'd be interested to hear more about how the algos detect the current market type & adjust their rules accordingly I was going to ask how the algo calculates correlation between positions but martyn was kind enough to create a whole playlist on the concept - this made my day! 💛 I try to use the volatility of each position for risk management (as andreas clenow teaches) & martyn's concepts could be a fantastic addition to it I'm stunned by how sophisticated (but still probably not over-optimized) his systems are. He's truely a role model for us rookie traders
Where can one access Martyn's correlation data? He should code a correlation indicator for manual scalpers as myself to utilize. Great interview. Great guest. Always my fav tube/podcast. Keep up the great work Etienne.
I think usually technical indicators on trading platforms say something about a single market, while this correlation is data about multiple markets. However I think TradingView has a correlation indicator, which measures correlation between one market and another. But what I think Martyn is talking about is when ALL markets (or at least the ones he's looking at) start to correlate (which could mean a general market crash or hype). I'm a (hobbyist) algo trader myself, and correlation between 2 markets is still something that can be very useful but you need to pick your pair carefully. For example 2 markets can be 'naturally' correlated, for example 2 companies that are in the same business and maybe even owned by the same parent company. You don't even need to know the specifics of all that, just do a correlation analysis of the markets and find a pair that are highly correlated. When that correlation weakens, chances are high that one is undervalued or overvalued.
@etinne Can you put me in touch with this guy? I love how he has his portfolio management system... it's exactly what I am trying to build for my algos. I'd like to see about collaborating with him on a dash that manages algo portfolios.
Great interview with nice insight into how to get the best performance out of a portfolio of algo's, definitely something of intertest for me as an early stages / beginner algo trader with a few different strategies running. As far as using AI for algo trading there has been a fair amount of research on this and from what I understand is that AI is much better at managing your portfolio rather than building strategies. Dr Ernest P Chan from predict now AI has done some really good work in this space, he may be a good person to interview for an in depth discussion on the future of algo trading.
Hey , I am new in trading .I want to be a full time trader. I am working on it hardly but I badly need a mentor to make my trading skill a sharp knife. Anybody please suggest me a good mentor to level up my skills.
Yes, it's very easy to get an insanely overfitted machine learning model using market data if one doesn't realize just how insanely noisy that market data is. But it can still be very useful if you know what you're doing and how to deal with all that noise. It's actually my main approach in algorithmic trading now, but it took some time to come up with machine learning models that can be trusted in that area.
No dude... this is not only for also traders, this is hard to understand if you only glance over this content. It requires a deep desire to understand otherwise you never will.
I occasionaly check in here. Most guys are discretionary dudes talking about bullshit mindset rules, this guys approach comes close to how real quants handle their business
Excellent material. I am developing my own algo for 4years and I clearly can tell that this lad knows what he's talking about
Spent a significant part of my life programming and a decent portion of that was automation in one form or another, started programming trading algos about 3 months ago and going strong rn, can't wait until its "perfected" 💪
Is it profitable yet?
Did you test it?
Great interview. It's always good to hear about the genuine algo traders and their techniques.
When a trader has legit track records, it is definitely worth to listen to. Rather than wasting time listening to someone but has no track records, smells like a marketer than a trader
This was encouraging to see. I’m currently in the process of building my portfolio and plan to be using them within the next few months. It was helpful to know that there are others out there building their basket of EAs in a similar fashion. Great video, thanks.
I love Martyn… specially how he overthinks everything.
TRADE like a machine in the building!!
Nice interview. I got funded at Darwinex a few years back, it took a rather long time. Perhaps because I didn't understand their home made criteria well enough, they're non-standard. Their broker platform is very good with low spreads. The profit split is very low compared to FTMO etc. As for multiple strategies, another option is to only use one low risk algo at a time on one asset. Then the correlation problem goes away. I've looked for such controling software with no luck. When it comes to predicting the market with AI, the big hedge funds have had zero success even though they employ dozens of PhD data scientists at a time. The problem I guess is over-fitting, you can fit historical data very well with all those parameters, but as the market is time dependent and never repeats the prediction power is low.
Great interview, as always!
In case of a follow up, I'd be interested to hear more about how the algos detect the current market type & adjust their rules accordingly
I was going to ask how the algo calculates correlation between positions but martyn was kind enough to create a whole playlist on the concept - this made my day! 💛
I try to use the volatility of each position for risk management (as andreas clenow teaches) & martyn's concepts could be a fantastic addition to it
I'm stunned by how sophisticated (but still probably not over-optimized) his systems are. He's truely a role model for us rookie traders
I like your interviews. I hope you will interview Bob Volman who wrote Forex Price Action Scalping. His books are great.
Thank you !! Good information and interview
Glad you enjoyed it!
Commodore 64! Had you down as a Speccy guy like me, Martin.
His tutorials (in Darwinex) are great as well!
Where can one access Martyn's correlation data? He should code a correlation indicator for manual scalpers as myself to utilize. Great interview. Great guest. Always my fav tube/podcast. Keep up the great work Etienne.
I recently found one on a indicator site. I'll check it out and see if its legit before i point you in the wrong direction
@@minimal2224 that's great. I've been flipping through indicator libraries looking for something.
@@minimal2224 lmk the site if it proves useful.
I think usually technical indicators on trading platforms say something about a single market, while this correlation is data about multiple markets. However I think TradingView has a correlation indicator, which measures correlation between one market and another. But what I think Martyn is talking about is when ALL markets (or at least the ones he's looking at) start to correlate (which could mean a general market crash or hype).
I'm a (hobbyist) algo trader myself, and correlation between 2 markets is still something that can be very useful but you need to pick your pair carefully. For example 2 markets can be 'naturally' correlated, for example 2 companies that are in the same business and maybe even owned by the same parent company. You don't even need to know the specifics of all that, just do a correlation analysis of the markets and find a pair that are highly correlated. When that correlation weakens, chances are high that one is undervalued or overvalued.
@etinne Can you put me in touch with this guy? I love how he has his portfolio management system... it's exactly what I am trying to build for my algos. I'd like to see about collaborating with him on a dash that manages algo portfolios.
Great interview E I know nothing about computer programming but it would be funny to learn and turn a profit
Quality material. 🙏🎉
As a statistician you've made my day!
Amazing interview - Thank you!
Excellent interview.
We let the data, not the intuition or instinct, to guide us. - Jim Simons. The best trader/investor/speculator whatever you wanna call it, ever.
Great point!
Great interview with nice insight into how to get the best performance out of a portfolio of algo's, definitely something of intertest for me as an early stages / beginner algo trader with a few different strategies running.
As far as using AI for algo trading there has been a fair amount of research on this and from what I understand is that AI is much better at managing your portfolio rather than building strategies. Dr Ernest P Chan from predict now AI has done some really good work in this space, he may be a good person to interview for an in depth discussion on the future of algo trading.
Amazing!
Hey , I am new in trading .I want to be a full time trader. I am working on it hardly but I badly need a mentor to make my trading skill a sharp knife. Anybody please suggest me a good mentor to level up my skills.
look up inner circle trader
There's already a ton of research being done in Ai eg. Neural Networks. They tend to heavily overfit
Yes, it's very easy to get an insanely overfitted machine learning model using market data if one doesn't realize just how insanely noisy that market data is.
But it can still be very useful if you know what you're doing and how to deal with all that noise. It's actually my main approach in algorithmic trading now, but it took some time to come up with machine learning models that can be trusted in that area.
As a host, you speak very fast. And this is confusing for people who are not good listeners
Gem packed this one… wish Martyn was my next door buddy.
No beaches involved? I'm disappointed now...
No dude... this is not only for also traders, this is hard to understand if you only glance over this content. It requires a deep desire to understand otherwise you never will.
Best vs. Worse … speaker.