Search results

Follow a scientific framework and diversify. If you know how to code, that would even be easier.

Sometimes it's hard to stick with a strategy or system when the drawdown hits. But often it's good for you to evaluate objectively If it's a regime change or statistical blip.

Wow wonderful insights! You seem like an experienced discretionary trader (something I can't do for nuts). I'm a systematic trader and will probably try to integrate all the factors above and commit Occam razor.

Thanks for the summary! I'm trying out for some quant roles and may need to familiarize myself with Kdb+. haha. For my own system, I'm using a mix of SQLite and plain old text files for my low frequency since no concurrency is involved. Haven't fully explored Arctic but will try it out too.

Let's say if you got 2 positions (pink and green) - the possible issue is you may not know which will go up or down at any point of time. When you merge the 2 returns stream, the ensuing returns stream in blue has low to minimal volatility because of the inverse correlation. When you are able...

Hi all, couple of posts back. I noticed that some of you are using MongoDB (via Arctic) to store your time series data. What's the disadvantage/ advantage over say Kdb+ (Q) which seems like a commonly used DB in the quant space? At the moment, I'm not well verse Kdb+ but plan to look into it.

Monitor your portfolio volatility like a hawk and you would avoid risk of ruin.

Mathematically, diversification among uncorrelated positions with positive expected returns will increase risk adjusted returns. With that you can always leverage up to boost your returns. That's what Ren Tech, Citadel do all day. We can take a page from that.

Have different trading rules. Scale in and out of positions. That will give you smoother returns. In stats, there's always a distribution around the optimal parameter. Even if it's the optimal parameter, you wouldn't want to place all your faith in that single rule/parameter. You can read Perry...

I think they increase their allocation since 2 quarters ago. You can read the Sec filings. I believe they allocate most of them to their risk parity portfolio. I'm unsure but that's a speculation.

Bridgewater has 20% in gold now. Unsure whether it's their full portfolio but that's what they disclosed.

Can consider selling volatility.

+15% in usd. +12% in home currency. Max drawdown of 10%. S&P 500 is not a benchmark for me since I maintain a pretty low annualized volatility of 10% in my main strategy portfolio. I definitely wouldnt want my portfolio at >40% volatility which is what it is at 1 point during the crisis. And...

I'm unsure what you meant by less complex stuff. Python is more than capable for Machine Learning, Time series stuff. If you are referring to HFT, you probably need lower languages like C++ and definitely not MT4.

https://www.youtube.com/channel/UCOSvCJJG7PTH-jhp0phvlrg/videos ibridgepy seems to be most suitable for your purpose. I personally use ib_insync for my strategies. You may use the function, order_target_percent(security, <percentage>) in your script.

Volatility is easier to predict since it clusters. Momentum - less so.

Leverage should be tied to volatility/standard deviation of your portfolio. Maintain a cap and you should be safe unless its some crazy negative skew assets - which means it shouldnt be levered at all.

If it's once a month, you can do it manually. If not, you can always write simple scripts to rebalance it using python ib_insync, ibridgepy packages.

I rely on python for backtesting and deployment. R is another alternative for backtesting. I'm not completely familiar with mt4. But I have seen its C-based syntax before. To me, it seems tedious to code in that.

Hello, spoke to you about hedging forex risk in your thread. I realize it's cheaper to hedge in Saxo than IB. Something for you to consider :) Though requesting their REST-based API permission from them is a pain.