Search results

IMHO, volatility is the std dev of returns, but risk is something different. Risk is the std dev of the residuals of your “model.” When you have no model, then volatility and risk end up being the same thing since the residuals become the returns around the mean (or 0 if there is no expected...

As a person who is reasonably knowledgeable in AI/ML, let me break down why the OP is full of shit. Why do I deem it worth my time to do this? Because posts like this piss me off as they are obvious scams and give people who are ACTUALLY working on AI/ML a bad wrap. OK, thanks for giving us...

each jurisdiction, country, taxes income and capital gains differently. I live in Canada, and capital gains is treated very differently here — it is taxed at half the rate of regular income such as interest or dividend income. While regular income is taxed at 54%, capital gains are taxed at 27%...

It’s all about tax treatment of income. If you buy SGOV, you get interest income, which is taxed as regular income. If you buy BOXX, you don’t get any interest or dividends, you get capital appreciation, which is taxed as capital gains. It’s immensely clever and quite an amazing product. But, if...

Just write whatever is easiest for you, especially if you’re not a developer. If, and when, you hire a developer, they will rewrite it many times :) The reality is that code gets rewritten often. Do the simplest thing you can get away with now. When you outgrow it, rewrite. There is no point in...

junkone, I can meet up pretty much any day if we’re meeting during the day, say lunch? It looks like Oshawa is about 44 mins away from my house and I can certainly meet you half way. it’d be nice to get more people together, and certainly everyone is invited from my perspective, but I’m not...

As far as I know, there is no “group,” it’s just Trendguy telling people they’re not good enough to hang out with him :) I’m in mid-town, Yonge and Egg, FYI. I’m up for meeting or chatting anytime. For context, I have an automated system that trades about 40 futures markets on daily frequency...

Should I include screenshots of your texts. You’re trading a single “micro” futures contract and act like you’re the shit around here. Get over yourself man. The worst is that you’re actually destructive to the community by actively working to be exclusionary. Who gives a shit whether someone...

Yes, you trade one market using TA and chart voodoo and yet you don’t think anyone else is a “real” trader, as if using chart astrology makes you one :)

Junkone, what do you trade? I have an automated futures system that trades roughly 40 markets using the IBKR API.

AI is a bit of a loaded term. Let’s focus on ML (Machine Learning) instead as I think there is less mysticism around it. The key thing to keep in mind about ML is that at the end of the day, it is a set of tools to extract information from data. Typically, this is in the form of fitting a...

Anyone up for meeting up this week? Friday perhaps?

In terms of minimum size, keep in mind what models like XGBoost are doing underneath. On each iteration, they are building a decision tree on your data. So, if you are using a max depth of 1 (for illustration here), which means one split on a single feature (pretty much the simplest model...

How deep were your trees in XGBoost? Also, I know that some implementations don’t respect a minimum split size, which is going to produce more random results.

I would love to meet-up with other traders in Toronto. I’m in midtown but can travel to wherever to meet.

When creating an ML model, assuming it is a supervised learning regression model (which is what you’re looking for in this particular case), you model the relationship between independent and dependent variables. Think back to algebra … y=f(x) … this is literally what ML is doing, it is...

Your model is WAY too complex. 4 layers with 50 neurons each is over 10K parameters. How large is your dataset? You’d want to have thousands of data points for each parameter due to large noise levels. So, unless you have over 10 million rows of data, your model will produce pretty random...

GAT, That's an awesome blog post. I haven't done much with the impact of interest rates yet, but I've started to think about other "conditioning variables" (volatility mostly at this point) and it's great to see how you approach the problem. Just to confirm my understanding, you use your two...

I realize my reply above does not answer the original question (arguably it doesn't even touch on the original question). But, the original question got me thinking about using correlation info for portfolio construction (which is pretty close to the spirit of the original question at least)...

Interesting question. As mentioned by other posters, it's a tough one. There are some papers out there suggesting that reducing risk (volatility target) in response to increased correlation between markets produces better results -- I believe that is in one of the Baltas papers. I've not...