>non CS related >going to build a predictor based on pic

>>1750628
what the fuck are you talking about

what kind of predictor
how the fuck is it based on this pic (is that a dali?)
what is emergence? link me

>>1750710
the painting is called
Dream Caused by the Flight of a Bee Around a Pomegranate a Second Before Awakening

was talking about the random flight of a bee

emergent behavior
https://www.youtube.com/watch?v=ctMty7av0jc

stock/options predictor

>>1750735
what kind of predictor?
bayes? ANN? KNN? decision tree?

>>1751060
maybe some hybrid one; i don't know yet. i saw something about finding the path of a fly that was interesting. i think I will play with one stock and try to have it statistically beat the market, so i'll end up losing some but in the long run i will stay ahead. it will text message me to buy and sell so i can manually set the orders.

>>1751138
i've been looking into using a neural network to do this sort of thing but there don't seem to be many resources that mention people's successes so i'm not sure what to think

>>1751146
hedge funds are already sinking millions of dollars into this shit, good luck to your little hobby

>>1751146
NN requires a lot of data to train--5 million+ observations. I implemented a hybrid time series model that was used to predict traffic, and it performed so much better than the linear model.


im having a hard to getting the data. i would also need almost real-time ticker data, which seems expensive for public data. i have scrapped google finance for data, but that probably won't be viable in the long run since it's by the minute. i don't plan on doing high-frequency trading. i'll have to trade slow but with more complicated models.

>>1751157
you are right it would be a hobby

>>1750628
>>non CS related
>>going to build a predictor based on pic related
>>/sci/ tell me about emergence
>>that emergence converges into a wave

cool story

blog it

>>1751190
i'm thinking to have it tweet the predictions