[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/biz/ - Business & Finance

View post   
File: 56 KB, 512x512, gwwojak.jpg [View same] [iqdb] [saucenao] [google]
30055123 No.30055123 [Reply] [Original]

After months of huge money losses experimenting with different trading algorithms and strategies, I have finally done it. Almost fully automatable, so far backtested on 6 straight months with >80% profitable days and averaging at about 300% monthly gains.

I only have a few thousand dollars left at this point, but it was all worth it. Maybe I will share my secrets with you guys in a couple of months after I make it.

>inb4 backtesting doesn't prove anything
If you're one of those retards who are telling everyone this as a blanket statement, pls kys.

>> No.30055369

>>30055123
Congratz anon, make sure you test your algos for all market cycles. It's easy to make a strategy that works in a bullrun but most get absolutely rekt during bear phases. Only testing the past 6 months will give you massively skewed results unfortunately.

>> No.30055680

>>30055123
which API are you using to automate your trades.

> t. dev thinking about doing the same

>> No.30055835

>>30055123
How'd you get started? I'm trying to do something similar with Python

>> No.30055873

>>30055123
Just hold... it’s really that simple

>> No.30055921

>>30055123
overfit

>> No.30056145

it's true, backtesting doesn't prove anything

>> No.30056546

>>30055369
Thanks. Yeah, I did keep that in mind and will keep back testing it further. I don't expect it to work this well during a bear market, but my logic tells me it should still be pretty good.

>>30055680
Don't want to reveal too much, but can just say I'm probably not doing stuff in a very conventional way and mostly just using the exchanges' APIs directly.

>>30055835
Literally just started learning python a couple months back for this purpose lol. Just googling around and shit, but started out already pretty solid in C.

>> No.30056722

>>30055123
How much for the src

>> No.30057222

>>30056722
Ask me again in a few months lol

>> No.30057679

>>30056145
Ofc it doesn't 100% prove anything, but if you know your own code and what assumptions you're basing the trades on and literally just use your fucking brain when interpreting the results then they can be pretty good indicators of the future. Literally all you have to do is to not be delusional.

>> No.30057848

I'm thinking of doing some neural net training with TensorFlow. I don't have great hopes, but I wanna learn a bit about them anyway, so...

>> No.30058204

Can I get historical market data somewhere or should I just start scraping exchanges? I would give all my curry for all historical 1 hour candle data of cryptos of my choosing

>> No.30058585

>>30055123
> hay guys my boy made money in this clown world bull market!

So did my ten year old cousin. Talk to us in a few months...

>> No.30058641

>>30058204
Coinmarketcap, coingecko, trading view, every exchange ever?

For someone who is so confident they can be rich with some historical indicators you sure seem to know literally nothing about this!

>> No.30058982

>>30058641
Coinmarketcap and Coingecko have _daily_ data for easy export. I was wondering if I could download a JSON or CSV of _hourly_ data somewhere or do I need to scrape it off from the small chart widgets, day by day.

>> No.30059535

>>30055123

Were you a successful trader prior to your bot? (meaning: did you just automate your existing trading strategy or come up with a new one altogether?) Is it a short term intraday trading strategy? Which tech indicators do you rely on?

>> No.30060122

anon give me a step by step tutorial to the utmost degree you are willing to reveal it.

>> No.30060410

>>30055123
Wait... are you saying that you were able to create an automated bot program that can buy cryptocurrencies for you and then sell them for PROFIT... in a bull market?