Search results

I don't really distinguish those cases the way that you do. Especially with respect to the thread. When people on the forums say that backtesting is a waste of time. I think of backtesting as an integrated process, much like R&D. from that perspective, I don't see that process as a waste of...

Since you are on the topic of ML. Few in ML would approach testing in the "tweak/test/fail, tweak/test/fail, tweak/test/fail, tweak/test/success! " approach to begin with. There is something called cross-validation and generalization in the field that is very well known, and is the antithesis of...

Thanks for sharing those. Can't really speak to the UHF, as I don't have much experience there. Something like a pattern though, I would think would be amenable to testing. I get the part about altering the outcome by impact, but I wonder how you arrive at the analysis that the pattern yields...

Backtesting is absolutely a research tool. If it helps, remove the word back, and just think about testing. You should run tests over data to draw any conclusions about it. I still have yet to see someone describe a (legal) edge that would not benefit from some sort of quantitative analysis or...

I think we all grapple with these thoughts over the years. Most of introductory time series modeling requires conditions of stationarity in order to be able to estimate and predict that properties like statistical moments can be reliably estimated at some time in the future. Something like a...

I actually see that as using a bayesian approach to statistics. Maybe not as formal, but still utilizing some form of statistics there. I can't say for certain how you formulated your priors (E(u_t+1|earnings) = -2%) > E(u_t+1|earnings) = +2%), but again, though it may not be formal, I would...

I would say that it's when you 'think' you have a true edge that lots and lots of testing begins. You are trying to find out where it could break and robustness, etc.. before executing it. By looking at lots of scenarios, you get a better expectation of variations in the outcome, so you know if...

Here's some results of a system that was praised and written about in various blogs right up until about late 2014. The indicator was first revealed in 2009. After 2014 we didn't see much about it anymore in the blogosphere, as the system performed poorly since then. This is an older blog...

It comes down to whether or not rules can be coded. ML can replicate pretty much anything you want, given that the input/output decision/response space can be mapped to a data set, but unfortunately, not in a plug and play manner. I still try to understand people who adamantly believe that it...

Worse than pointless... it's very misleading. Like showing a single outcome of a coin toss equaled heads, as if that implies something. Some of you TA guys really have to get out of the dark ages. To add to that. Find some way to quantify that setup, so that you can somehow distinguish a...

You don't prove or disprove anything by showing anecdotal chart examples. You start by tabulating contingency tables and probabilities over large sets of examples . :banghead: At least Bulkowski did some of that for you. Start from there, not selectively choosing charts.

I recall reading that they use bitmap to values of images of the charts, which are simply a matrix of (binary for black and white, 256 for RGB) values presented as input features to the learner. It is common to flatten the image matrix into a 1 dimensional row vector for each input observation...

Here's a slightly alternate version of the paper posted. https://arxiv.org/pdf/1912.01708.pdf I'll take a stab. My initial takeaway, is that he is saying that markets can be favorably manipulated by large enough short term players that implicitly collude, and he explains some possible methods...

Just looking at that paper raises several flags, and makes it not really worthy of continuing (other than academic theory)... Some examples 1) Use 1 year of data??? 2016 that's great for the one year of data. Too bad next year may be nothing like it. 2) RMSE is kind of a useless metric to your...

A moving average model is not necessarily bad or good, it is just a model. One could argue that how you use a model is more significant than the model itself. One thing I realized is that a simple moving average (from TA world) is useful to beginners since it's a very simple model for neophytes...

Very, very informative book for those of us who have always felt like outsiders. Amazon sold out on release day, and I had to run to a local Barnes to pick it up. But, I didn't put it down and read it in one sitting. Really gave me a big sense of respect and belief for Simons. It wasn't always...

From a mathematical and signal processing perspective, Left has a higher signal to noise ratio... Right has a steeper slope... From a financial engineering perspective, lower variance (left) is often preferred. Lower reward, but lower risk. Think portfolio diversification. Many traders like...

So, the GA is likely generating random continuous parameters with a granularity less than your step size of the brute force generator. The fact that many of these do much better, shows it is highly sensitive to small parameter changes and thus not likely to perform well out of sample. Not only...

Genetic Optimizers usually work by randomly seeding and selecting populations. You should not expect results to be exactly the same each time. If you truly want to compare apples to apples, you need to brute force all of the possible outcomes. That will of course, truly over-fit the best...

OP. Interesting work. What kinds of research of his did you find practically useful? Just looking at the Hudson presentation slides, shows Equal Weight TW performed much better than all of the competitors, including HRP. If you leverage up HRP, you get similar results. I.e. same Rwd/Risk ratios.