/biz/ - Business & Finance

>>21597575
well it's not really 100, the way im using it normally is with "full range" on so for 5 m 5D it would use all of the candles for the entire data set. you can set it to less candles though like pic related.

i wouldn't really call it an exit for KODK, the way i look at it is that you'd scale in to a position through 2+ SD, and exit when it regressed to or near the mean. so you'd short some KODK on the way up at each level and cover when it dropped down into profit.

but also, like i said, i think it works better on a more "normal distribution" kind of stock that isn't going to make crazy moves like that out of nowhere. i guess it would depend on what the news was and whether you think that would merit the stock continuing in the extreme direction or not, or of course you could short fewer shares for less risk depending on volatility, or even just wait for momentum to reverse, say on the macd or some other momentum indicator (like waiting for the RSI to leave being overbought and probably going to 1m instead of 5m for the time frame to have more granularity).

you're right though in that im still experimenting with what data set to use for what time frame, like day or short-term swing trades. i've also been taking screenshots of stocks on different timeframes at different points of the day to see how the channels change. im thinking 5 days worth of data is probably pretty good for day trading currently for what levels you want to enter at and to get a better sense of the overall trend. like today was a great dip buying opportunity on AMD for example, i made a bit over a dollar per share trading it today.

>>21594897
also thanks for pointing this out. basically by default it calculates using the entirety of the data set. but if you set it to "no" then you can set how many bars it calculates by. scalped a little at the end of the day here and it was actually pretty neat, also tripled the default value. might have to experiment with this as well for scalping, could be cool.