Twitter volume predicting price movements?

[cryptofan2017]

Would twitter volume on a certain altcoin provide a usefull signal for price movement? As, especally small cap, altcoins seems to be still heavily moved by sentiment twitter volume could be a usefull signal. I saw a twitter volume analysis on litecoin at www.altcoinanalytics.com that seems to show some support for that. What is your opinion?

[1715308121]

1715308121

[1715308121]

1715308121

[1715308121]

1715308121

[e-coinomist]

Would twitter volume on a certain altcoin provide a usefull signal for price movement? As, especally small cap, altcoins seems to be still heavily moved by sentiment twitter volume could be a usefull signal. I saw a twitter volume analysis on litecoin at www.altcoinanalytics.com that seems to show some support for that. What is your opinion?

How would you measure "twitter volume"? Counting tag occurances? Calculating in the reach of the account mentioning them?
Be aware people can buy services on each and every social network. Your numbers might get gamed by social bots.

[CoinCollector006]

Yes, there are a number of papers in the area which have shown promising results and this has been on my to do list for a while.  The devil is in the details of course.

Here are a few links in case you are interested in digging deeper.

Algorithmic Trading of Cryptocurrency Based on Twitter Sentiment Analysis
http://cs229.stanford.edu/proj2015/029_report.pdf
This paper predicted the hour to hour sign change of the bitcoin price using Logistic Regressions, Naive Bayes and SVM.

Predicting Fluctuations in Cryptocurrency Transactions Based on User Comments and Replies
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4988639/
This paper uses forum comments to attempt t predict price changes of Bitcoin, Ethereum and Ripple

PS these papers assumes a fairly high degree of statistics, machine learning and programming expertise and faculty with data hacking.

I agree with the above poster that some of the data may be gamed by bots.

[左]

Some coins gather a lot of people to do promotion, some coins don't do bounties at all.
The coins with bounties are obviously going to be represented on twitter a lot more, but doesn't necessarily mean the project is more popular or promising.

[iron talon]

Would twitter volume on a certain altcoin provide a usefull signal for price movement? As, especally small cap, altcoins seems to be still heavily moved by sentiment twitter volume could be a usefull signal. I saw a twitter volume analysis on litecoin at www.altcoinanalytics.com that seems to show some support for that. What is your opinion?

How would you measure "twitter volume"? Counting tag occurances? Calculating in the reach of the account mentioning them?
Be aware people can buy services on each and every social network. Your numbers might get gamed by social bots.

This can give an idea of market and its society, but you can't say this is a valid data to find out what's gonna go up or down as coins. Relying on this data, would be amazingly wrong.

There must be some bots scanning the twitter tags etc.

[monkfishy]

Yes, there are a number of papers in the area which have shown promising results and this has been on my to do list for a while.  The devil is in the details of course.

Here are a few links in case you are interested in digging deeper.

Algorithmic Trading of Cryptocurrency Based on Twitter Sentiment Analysis
http://cs229.stanford.edu/proj2015/029_report.pdf
This paper predicted the hour to hour sign change of the bitcoin price using Logistic Regressions, Naive Bayes and SVM.

Predicting Fluctuations in Cryptocurrency Transactions Based on User Comments and Replies
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4988639/
This paper uses forum comments to attempt t predict price changes of Bitcoin, Ethereum and Ripple

PS these papers assumes a fairly high degree of statistics, machine learning and programming expertise and faculty with data hacking.

I agree with the above poster that some of the data may be gamed by bots.

Interesting, thanks

[pey]

Yes that can work but it may sometimes fail as well. Taking into consideration only twitter volume clearly is not a good idea, you should consider many factors.

[MatrixMining]

Twitter volume can help, but it can be highly influenced by bounty campaigns. It's thus probably better to check also Google search volume for example and consider all other factors like upcoming development news etc.

[imperatron]

You can find some good hyped coins on twitter, but thats not the best way to do some researches!

[e-coinomist]

Here are a few links in case you are interested in digging deeper.

Less scientific but allready up and running https://twitter.com/cryppadotta/status/913062920690339840
Dotta (that mupped lol) provides some data since a while. The selection of twitter accounts has been done manually. Those twitter people are not unrelated, some of them are forming "teams" to peddle their own agenda, use with care.