Search results

I work in data science, i understand what you mean. One could also use k-fold cross validation for that matter. However, when the (hypothetical) equity curve is smooth this would essentially mean the same thing, imo

So if the method is tested on 15 years of data, you would wait for 7,5 years? Not practical imo. Better slowly advance with real money

To provide reasonable answer you need to give more details about your ‘train set’ Imo the number 1 question for possible overfitting would be: do you have reason to believe that your future dataset (trades) will be different then the dataset you tested up on. know also that it’s by...

Guy in my street has an urus (lamborghini), but not that impressive at all for the (high) price. He'd better bought a Marauder. It has nice buttons in the front:

I do not know if this true-once-daily dosing would be the main driver for succes, but once-daily dosing does already exists. I personally have some adhd (who doesn't nowadays) and use a form of Lisdexamfetamine, Elvanse/Vyvanse. You can take this once in the morning. Huge advantage is no...

I think we have to wire $300 'banking transaction costs' into your nigerian account to get acces to the trillion-system?

Haha, more like, 0.75 isn't always more than 0.25

This is indeed a bit surprising. Considering myself, as an 'economist' who morphed into a 'data scientist' I found the field of the real data and programming people very 0-1 thinking. Difficult to explain, but in general programmars are often very 0-1, while reality is more fluid, unclear and...

I did some analysis on the thesis of seasonality, could be interesting to share. So this is only for data on spy, 1993-now. I've used weekly returns, so numbers on the axes represents weeknumbers. So first a comparison of the distributions of weekly returns, per week. Each line shows the...

Interesting, will look it up! Was unknown with this field. With to post i was reffering to the 'standard' ml models, svm, knn etc, fwiw

With this kind of analysis it is always wise to look at the distribution of returns instead of averages, since outliers can distort averages. Then also look if the trend (year wise) is constant. If that's the case it could mean something. (Just be careful to jump to conclusions)

Machine learning is just modelling an outcome to variables. An outcome can be lineair or categoric. Machine learning is just modelling these outcomes from your given variables, and "the machine" returns you the best model. Basically when correlation is high your model will be well fitted to the...

You didn't mean in 1 day, right? ... What a gap!

Have you guys tried DAN? Should be fun for you

Great call Vanzandt, you do it again, again and again :thumbsup:

I think this is only possible if you are into a few instruments and know the prices by heart. Otherwise indeed numbers (with filters) are better and faster instead of graphs imho. In the end focussing on what idea/strategy to trade is what its about, not thze rainybows charts. Unless one wants...

Short term it is. You are right.

Well you claim trading from 20 to 40 and from 40 to 20 is the same ROE. I would say half. Further you claim percentages don't matter, only absolute numbers. So I'm curiuous how you can make $50 from shorting from a price of 40. (No leverage, 1 stock and 1 trade). *and no joke the 40 is measured...

How can i gain $50 from shorting that - without leverage and 1 stock?

The best reading productivity hack is imo not to read as much as possible in a given timeframe, but to just read less (nonsense) like my reply;)