You've Been Warned

AI price prediction for where BTC is going.

Take your profits now. This is your only warning.

Attached: tokenmetricsgraph.jpg (1773x730, 104.06K)

Other urls found in this thread:

tokenmetrics.com/
help.tokenmetrics.com/en/articles/4147995-how-token-metrics-uses-ai
twitter.com/SFWRedditVideos

Historic Accuracy

Attached: tokenmetrics.jpg (594x637, 60.57K)

Which bot is it? Neurobot?

So far it was able to predict the March flash crash accurately down to the lowest point you could buy BTC for, Jul-August false bullrun.
Currently it thinks we're headed down.
If you're still bagholding after this, it's your own fault.
The time to buy crypto was in March and April.

Onky 45% off last november!
Lol. Just using the current price at t as prediction at t+1 would have been a better predictor. Non STEMers are not fit for trading.

He's right, i ran some tests, did some analyzing and can conclude the same.

Attached: btc dying.png (854x431, 49.43K)

You're literally pointing to a single anomaly that lasted for a few days. That's what you call manipulation and not organic growth.

I see multiple points that are off by more than 20%. You know how hard the market will REK you with that accuracy?

Unless you're a day trader or have a leveraged position, why do these spikes bother you? The macro trend has always been on point.

I also see alot of points where the direction is wrong. So you would go long based on the prediction and would get raped or vice versa. Muh AI. Its probably just a simple timeseries algo.

Going long is for leveraged losers who gamble their wealth, this is designed for long term investors in mind.

Then give more specifics of the model and I will invalidate it for you. I have a PhD in Econometrics and have been a quantitative trader for over 5 years.
You cannot predict the fucking market looking at the price alone. Period.

I will discuss the econometrics of this model if you share them with me. But just for context: these faggots also use TA. Meaning they draw memelines and write horoscopes. Kek. But AI sounds good dude!

Attached: Screenshot_20200811_142255.jpg (720x1520, 298.74K)

It's this one tokenmetrics.com/

Article
help.tokenmetrics.com/en/articles/4147995-how-token-metrics-uses-ai

> we also provide 30-day price predictions through our Multi-layer Perceptron Model (MLP), also known as a feed-forward artificial neural network.

There are more articles though if you want to dig in.

96%? pleb
i have a 8 digit iq and my bot has 150% accuracy predictions, it says it's going to 100k eoy with a 1337% confidence

Thanks. 10 seconds in I am already convinced of how shitty their model is:

We simulate 100 different weighting schemes, back-test each one, then we pick the best performing model in terms of accuracy, to then create new generations.

So with each generation, our models increase in accuracy.

We repeat this process over 500 times, resulting in picking the best performing fundamental analysis grade model from 50,000 different models.


Any practioning quantitative trader knows THIS IS BLASPHEMY. This does not work. This is university professors looking at the market.

Anyone remember this?

Attached: 1576606035305.png (1172x578, 194.11K)

Bears are fucking delusional now

Attached: 1511215278240.png (600x842, 141.96K)

Im actually laughing at how dumb they are. That backtesting approach shows 100% that they do not understand how markets work or what a good quantitative model is. Thanks for the laugh op.

Honestly IDK what will happen but I was able to time the market correctly back in March so I got that going for me.
Even if there is some truth to it, I took profits on fundamentally weak projects.
Still holding BTC/ETH/DAI (30/30/40) %

Thats fine. Just dont go around "warning" people with such a shitty model. Who are you to warn people? The model is way worse than S2F by planB which is also a shitty model.

this scares me

Is this called "overfitting" or is that the incorrect term for their error?

fuck me. i have money in april... but didn't buy. Now my life goes downward without money. i don't know what to do .

>11k stablecoin

Based

Can you give some analysis on the state of crypto market? Are you invested?

I have an undergrad level knowledge of econometrics.

FML. With the corona still up i thought another dip was coming. Fuuuuuuuuck! 3 years later and missed the train again.

There will be no details shared but I do have a method for predicting movements in bitcoin (larger movements, not day to day - that is possible but it takes too long). It is a method I have been using for about one year and it has successfully modeled the market in that time. It was applied retroactively and mapped correctly to each prior year and the key trends.

You have no reason to believe me but I'm not here for attention and won't reply to this thread. What you need to know is that BTC will continue to crab in the $10k-$14k range from now until around Mid-November/Thanksgiving. It will not go lower than $8500 and if it does spike that low it will recover very quickly - that spike would be a short-term anomaly. Around mid/late November it will start to climb and will end the year around $15-$18k. These numbers are rough for reasons you'll understand in a moment. The important detail is that it will end up 2x on the year.

Beginning immediately after the New Year, Bitcoin will take on new life. it will climb around 12.5x during the year 2021, peaking in December. The higher it closes 2020, the higher the peak will be EOY 2021. Sell around Christmas or in the days following.

Bitcoin will proceed to dismantle its new ATH beginning after New Year's 2022 and will decline and stagnate for all of 2022 and 2023, very similar to 2018/2019.

You are welcome.

Attached: 1562132699324.gif (400x400, 85.8K)

Easier to criticize, let's see your model.

>for reasons youll begin to understand in a moment

Attached: 20200805_181226.jpg (1079x565, 80.99K)

>BTC going to shit
ok lmao
>ETH, LINK and ADA doing the same shit
I doubt it 500%
>Parity at $1000 after crash with ETH shitting on BTC 2 months after
It will happen

Attached: 2B434882-3657-4519-9601-A336E0045519.jpg (225x225, 9.92K)

Where's the rest of this chart? I don't wanna see just August. We are almost half way through it.

Please accept me as your student, sensei. I have nothing but a degree in statistics. I wanna build a custom quantitative indicator. Please let me know where to start.

I know Chainlink is amongst the more popular coins here.
So here is the chart for it. Looking good.
IDK how you can be more accurate than this.

Attached: link.jpg (1794x598, 72.29K)

Is this back testing or real?

It only gives price prediction for 30 days in the future, anything more than that is a literal crapshoot.