Search results

I'd delete it if I could.

and look who he references... Elder. Not a very good conclusion for an academic paper. Also, take a look at the results -- In sample buy and hold naive learner trounces his NN, yet he completely ignores the comparison OOS. Conclusion, not a very good paper to source IMO.

I know I'm fishing here, but.... Please post if you know of one, or even better if you've built one let me know. I know matlab has particle filter, but I'm specifically looking for unscented and non-matlab OS implementation.

Seeing as I discovered the source of your 'hidden' treasure, I'll take that as a complement. And, btw, who said rhino success was a bad book, there were many truths in it. Develop a skin like a rhino and gravitate to the top.:D And truth be told, I like your book better than brooks...

Optimal f sizing makes a potentially bad assumption about underlying distribution based on historical data. One or two fat tails can destroy you based on that assumption. There's the rub.

One thing I can say about amzn, and if you go back on the chart to analyze, is that they will go sideways just long enough (weeks to months) for bears to cry uncle in desperation. What looks like an easy target is not always the best. Especially if it is fighting the rest of the market. If...

So is beals book on hypnosis and the cosmic mysteries of the universe working out for you? Reminds me of an old sales book they used to push,Rhinoceros success.:D

Don't ever forget that his book, conquer the crash, get out of the markets, and run to the hills and eat canned beans was released right at the bottom of 03. Talk about bad timing. He has a history littered with worse than bad calls. Flip a coin, and you'll likely do better. Sometimes...

Hi Terry, The paper looks interesting. I haven't read too much yet, but-- there is a thesis I think will interest you. When I track it down, I'll PM you. It has to do with using hidden markov models, whereby each transition and emission probability is based on a mixture model that gets...

This is interesting, because I've found some different conclusions. First off, not all automated strategies exploit a static parameter based rule. Some adapt dynamically. 2nd, I find it hard to believe you stumbled upon 80% success rate (is that hit rate only? what was theil and rmse) in...

wow, amazon sure has a history of blowing shorts out of the water. This time actually might be on merit for once.

"Eventually humans will be needed for little except to reap the rewards of robots working for free." For a king of shorts, you sure are optimistic.:D I don't doubt the part about robots being productive, the part about everyone reaping rewards though...

Maestro! Finally, a sound thinker has arrived. Hope you are doing well, and the board has sorely missed your sage insights! Look forward to hearing more from you and glad to hear you are doing well.

AyeYo, piece of advice. When something smells like bs and needs to be obfuscated in non-nonsensical gibberish, it usually is. You're thinking correctly to begin with. Don't dilute your proper understanding by heeding the siren calls of the mad.

We also didn't get a 10 sigma drop in the last 20 yrs, go figure. Markets rhyme, not repeat. I hope you can at least agree not to fight the tape.

Not quite, and considering it's the highest weighted stock on Qs, good luck to shorts, cause it ain't looking bright. Granted, the markets been skyrocketing, but the same occurred in 03 as I mentioned in many earlier posts. And it lasted quite a while. Don't fight the tape... when will...

Speaking of winning at the machines. You guys need to check out an episode of criminal masterminds (title ?). There was a guy who QA tested video poker machines. Well, apparently knowing that ICs never generated truly random code, he was able to deduce the seed and modulo algorithm for certain...

Of course you can. If you ever take the time to empirically study simple probabilities, you would realize the market is a bent coin (from a binary perspective). Predict heads every day and you will beat a fair coin. Big problem that no one seems to notice is that markets returns are not...

That's an interesting approach. I think I'm still not understanding how you are formulating the problem description specifically enough. From my interpretation, all these methods are still classifying past observations into one regime or another. I am still missing how it infers the out of...

ok, so then you need to quantify it a bit more. If you are using cycles to identify dominant modes, then perhaps you can run a histogram of the cycles you record. If the histogram shows one or more dominant modes, you could identify the regimes by the modes you've identified. Mixture...