Google trends and the bitcoin price

[giarc]

I would like to discuss the level of search interest in bitcoin using data from google trends. Whilst I am a firm supporter of bitcoin and believe it will (with a little luck) go on to be a major player in the currency/commodities markets, it seems that the level of interest is in a bubble. This bubble in the level of bitcoin searches may or may not imply a bubble in the bitcoin price, it may imply a portion of the current price is due to a bubble factor whilst some of the gains are real growth.

Im sure few people would doubt that google searches are a strong indicator of the expected level of bitcoin price, this has been shown very clearly over the last few years, I have constructed a graph below to demonstrate the correlation.



If you look at historical search volumes and the ratio of these volumes to price it also seems that recently since late 2011 this ratio has been rising possibly showing the formation of a bubble.



Whether this is actually a bubble or a new paradigm is hard to tell, it could be that the old ratio was only correct for bitcoin when it was at a low market capitalization possibly due to the levels of demand from early investors and latecomers being at different levels.

One thing I would like to note though is that at the current time the number of google searches for "bitcoin" exceeds the number of searches for "bonds", it is also at 44% of the search volume that "USD" has which to me seems excessive at this time. Should a relationship between search volume and bitcoin price remain to be true then when this corrects the price should plummet to lower more stable levels.



This will be good for all long term holders.

I would be interested to hear some other opinions on this Look forward to your replies.

[XXthetimeisnowXX]

not sure about the bond correlation but the search indicator and bit price is interesting correlation, one that will continue for until people know about bitcoin, so for another year maybe

[giarc]

Im not saying that there is  a correlation between bitcoin and bonds, just that it would seem very strange for such a vast asset set such as bonds to be dwarfed in search volume by one currency. This to me suggests bitcoins search volume is unusually high, as search volume seems correlated to btc price it seems reasonable to predict a correction.

 

[jinni]

I think we still have a long way to go...

[RodeoX]

Some interesting data there giarc. Thanks for your contribution to the forums. 

[foo]

Im not saying that there is  a correlation between bitcoin and bonds, just that it would seem very strange for such a vast asset set such as bonds to be dwarfed in search volume by one currency. This to me suggests bitcoins search volume is unusually high, as search volume seems correlated to btc price it seems reasonable to predict a correction.

 

"Bonds" is an English word, "bitcoin" is the same in all languages. You would need to also translate "bonds" into all possible languages to make it a fair comparison. Also, Google Trends measures new interest, people who already know about bonds/bitcoin won't be searching for that word.

[Peter Lambert]

Im not saying that there is  a correlation between bitcoin and bonds, just that it would seem very strange for such a vast asset set such as bonds to be dwarfed in search volume by one currency. This to me suggests bitcoins search volume is unusually high, as search volume seems correlated to btc price it seems reasonable to predict a correction.

 

I would suspect the google trends data would be more closely related to the derivative of the price. As more people find out about bitcoins, the price tends to rise. Bitcoins are inflating at something like 12 % per year, so if there are no new people joining in the price will naturally trend downward. These two effects couterbalance each other, so at a certain level of money coming into the system the price can stay level.

[giarc]

Im not saying that there is  a correlation between bitcoin and bonds, just that it would seem very strange for such a vast asset set such as bonds to be dwarfed in search volume by one currency. This to me suggests bitcoins search volume is unusually high, as search volume seems correlated to btc price it seems reasonable to predict a correction.

 

I would suspect the google trends data would be more closely related to the derivative of the price. As more people find out about bitcoins, the price tends to rise. Bitcoins are inflating at something like 12 % per year, so if there are no new people joining in the price will naturally trend downward. These two effects couterbalance each other, so at a certain level of money coming into the system the price can stay level.

I did investigate this thinking that this might be the case however I couldn't find any correlation. I would suggest that most searches are by those who have already decided to invest but are trying to keep up with events and that those who search the most are likely to be major bitcoin fans.

[hgmichna]

Bitcoin price is a good predictor of Google Trends.

[nrd525]

Can you try correlating it with a delay?

If google search trends measure interest in bitcoin - and it takes people a couple days to invest money - then there might be a stronger correlation if you add a 1-7 day delay.

[dancupid]

"Bonds" is an English word, "bitcoin" is the same in all languages. You would need to also translate "bonds" into all possible languages to make it a fair comparison. Also, Google Trends measures new interest, people who already know about bonds/bitcoin won't be searching for that word.

In China bitcoin is also called 比特币. It seems the Chinese version is currently more popular in China:





http://www.google.com/trends/explore#q=bitcoin%2C%20%E6%AF%94%E7%89%B9%E5%B8%81&geo=CN&cmpt=q

[Kazu]

"Bonds" is an English word, "bitcoin" is the same in all languages. You would need to also translate "bonds" into all possible languages to make it a fair comparison. Also, Google Trends measures new interest, people who already know about bonds/bitcoin won't be searching for that word.

In China bitcoin is also called 比特币. It seems the Chinese version is currently more popular in China:





http://www.google.com/trends/explore#q=bitcoin%2C%20%E6%AF%94%E7%89%B9%E5%B8%81&geo=CN&cmpt=q

INB4 bond bubble.

[beckspace]

for bitcoin:


for 比特币:

[GeoRW]

I wouldn't compare bitcoins to James Bond 

[giarc]

http://www.google.co.uk/trends/explore#q=bitcoin&date=today%201-m&cmpt=q

Seems the search volume may be beginning to peak as I suggested in my post. It could be that this is a consequence of the price fluctuation rather than a mark of a  media interest bubble though, it will be interesting to see what the price does over the next few weeks.

The previous correlation between price rise and search volume would suggest an average drop of $40 based on this change in search volume.

Thats based on a correlation analysis only using search volume drops, price sensitivity to rises in search volume seems much higher than search volume fall sensitivity. For example a 1% rise in volume implies a $15 price rise but a fall in volume of 1% implies only a $7.20 fall, another hint that this may be a bubble or perhaps a modeling error on my part after all correlation does not guarantee causation.

[jinni]

http://www.google.co.uk/trends/explore#q=bitcoin&date=today%201-m&cmpt=q

Seems the search volume may be beginning to peak as I suggested in my post. It could be that this is a consequence of the price fluctuation rather than a mark of a  media interest bubble though, it will be interesting to see what the price does over the next few weeks.

The previous correlation between price rise and search volume would suggest an average drop of $40 based on this change in search volume.

Thats based on a correlation analysis only using search volume drops, price sensitivity to rises in search volume seems much higher than search volume fall sensitivity. For example a 1% rise in volume implies a $15 price rise but a fall in volume of 1% implies only a $7.20 fall, another hint that this may be a bubble or perhaps a modeling error on my part after all correlation does not guarantee causation.

This decline is probably because of weekend fluctuations. Just wait until Monday hits the Google Trend charts.

[Viceroy]

It's Monday and it looks like NOBODY is interested in searching for bitcoin anymore.  This is just a glitch, right?  

http://www.freeimagehosting.net/xk4kq

[Ares]

This happened to me once before, and for someone else it showed an all time high at the same time. So yes, I believe it's a glitch, or some irrelevant reset feature...

[BitcoinTate]

Yeah it is a glitch, I have seen it happen with several other searches over the past year.

[Viceroy]

what are you guys seeing?