For your purposes, you can just build a n-branch tree starting with the lowest correlation pair or something like it. However, i really think that before this exercise, you want to think of “what” correlation you are using. For long term strategies that’s a more important factor.
I agree with you totally. Just to clarify, are you referring to window size for correlation or using some other form of metric ? I am aware that there one can write other constraints still based on correlation to build different clusters. There is also a difference in using prices or returns and I find that returns seem to give me clusters that make more sense using domain knowledge. Actually this exercise is to run pysystemtrade with a reduced set of instruments to test various trading rules because the default set takes a long time with estimation on using shrinkage but I wanted a good representative set that wasn't picked randomly. However, building meaningful clusters is always useful in trading and has other applications.
Hi Rob, is there a YAML file or other where you store the initial & maint/overnight margins for futures instruments, or do you just check them before entering/adjusting a position? I've looked through several YAML files and Python scripts, but don't see anything.
@globalarbtrader, in the last comment here https://qoppac.blogspot.co.uk/2017/09/smart-portfolios-post-about-book-nn.html you say in terms of your risk allocation, 75% is in 'smart portfolio's, 25% in systematic trading. Why don`t you have 100% of your risk allocation in systematic trading given it has (I assume) a higher expected Sharpe ratio that the smart portfolios?
Someone just asked me exactly the same question on m y blog (Was it you? If so please don't do that!) Here's exactly the same answer: Basically trading is unsuitable for providing an income, as the returns you earn (essentially any profits over the HWM after tax) are lumpy and may be absent for years. As I rely on my investments for the bulk of my income it makes sense to put more money into dividend earning stocks and coupon paying bonds. GAT
It was me, apologies, I didn`t receive any confirmation on the blog post, so I wasn`t sure the post worked. Anyways, back the to question. Assume you have 200K to invest, systematic trading has higher SR, and you decide to put 100% of risk in systematic trading, targeting 30% ann. stdev. Let`s also assume the cash required for margin is 60% of your portfolio, which stays with the broker, and thus you have 80K in your bank account earmarked for trading. Is there something wrong or impractical with withdrawing certain amount every month from that 80K left in the bank account as a source of income?
Is that a mathematical thing or a psychological thing? Would a regular coupon from your trading capital hamper you in the long term?
The way I read this is by imagining a drawdown of 50% with 3 years to get back to the HWM. I wouldn't want to compound my capital down by taking a coupon in this scenario.