Topic: Search Volume as a Forecasting Indicator - page 2. (Read 4656 times)
makes sense
My take: Bitcoin enthusiasts are generally long on Bitcoin, and the general level of excitement/chatter/blagging/instant messaging on Bitcoin-related topics varies in response to the exchange rate.
This chatting/blagging/instant messaging/etc. creates an ongoing flow of bored/reluctant girlfriends/family members/friends/work colleagues who go on Google to find out what all that 'Bitcoins' [sic] nonsense is all about.
This chatting/blagging/instant messaging/etc. creates an ongoing flow of bored/reluctant girlfriends/family members/friends/work colleagues who go on Google to find out what all that 'Bitcoins' [sic] nonsense is all about.
lol
bitcoincharts has a percentage scale option, and uses the start date of the chart as 100%. Choosing the same start date where you find 1.0 on a google chart would produce the most accurate comparison.
The y-axis scale is irrelevant. I wanted to track how changes in search volume and in price correlate in time. It doesn't even have to be on the same plot, as long as time scales are same we should be able to see what's happening. 
bitcoincharts has a percentage scale option, and uses the start date of the chart as 100%. Choosing the same start date where you find 1.0 on a google chart would produce the most accurate comparison.
Google search volume has been mentioned over and over again as a relevant indicator. Presumably, a spike in search volume is a bullish sign. Alternative explanations go along the lines of search volume following the price, not the other way around.
So, where you got the data for the graph? Share your references, so anyone can verify the data reliability.
Search volume is available from google, e.g. https://www.google.com/trends/?q=bitcoin&ctab=0&geo=all&date=ytd&sort=0
the search volume index is charted at percentage scale, btw. you're comparing it to bitcoin price in linear scale and log scale.
According to Google:
Quote
In fixed mode, the data is scaled to the average traffic for your term during a fixed point in time (usually January 2004). In our example, 1.0 would be the average traffic of dogs in January 2004. If you chose 2006 as your time frame, you would be comparing data for dogs in 2006 to its data in January 2004. Since the scale basis (1.0) doesn’t change with time, you can look at different time periods, and relate them to each other. (Note: For keywords without a historical record, it may not be possible to establish a fixed scale).
It is not clear how they establish reference volume for keywords that didn't appear in 2004, but regardsless the scale is arbitrary and fixed. Anyhow, the values were close so I didn't bother including secondary y-axis, that is all.
What I see is a trailing indicator: price moves and then search
volume follows. In other words, not particularly useful.
It's only useless until someone starts a google search volume futures exchange.
Google search volume has been mentioned over and over again as a relevant indicator. Presumably, a spike in search volume is a bullish sign. Alternative explanations go along the lines of search volume following the price, not the other way around.
So, where you got the data for the graph? Share your references, so anyone can verify the data reliability.
the search volume index is charted at percentage scale, btw. you're comparing it to bitcoin price in linear scale and log scale.
The Bitcoin price is a good indicator of how the google search traffic will be.
Use it at your own discression!
Use it at your own discression!
If only we could short Google search volume.
proudhon, you are one handsome bull!
The Bitcoin price is a good indicator of how the google search traffic will be.
Use it at your own discression!
Use it at your own discression!
If only we could short Google search volume.
The chart tells me that the only way to make money from this correlation...is if one could buy/sell a futures contract on search volume for the "Bitcoin" keyword 
bitcoins are over-priced at the moment! 
yes, there is a correlation, but i don't think it's that easy to jump to any conclusions. the search volume is probably an aggregated value for the past 30 days, or even more. you don't know how exactly google does the calculation! the price doesn't have this kind of lag.
The Bitcoin price is a good indicator of how the google search traffic will be.
Use it at your own discression!
Use it at your own discression!
this graph lead me to believe the resent dump of coins was a really dumb ass move.
on second thought... maybe it was a needed correction
nvm ... still ... CONFUSED!
on second thought... maybe it was a needed correction
nvm ... still ... CONFUSED!
Google search volume has been mentioned over and over again as a relevant indicator. Presumably, a spike in search volume is a bullish sign. Alternative explanations go along the lines of search volume following the price, not the other way around.
Here you can see overlapped historical bitcoin prices from MtGox with Google search volume for "bitcoin". Conveniently, the "exact" search volume scale (as opposed to the relative scale) mostly matches the USD exchange rate, so it's all on the same unnamed y-axis.
Bitcoin price data are daily average, while google data are unfortunately compounded weekly. I have therefore placed google data points on the 4th day of the week.
Also shown are the same data in log scale so no detail is lost in the valleys.
What do you think? To me, the conclusion is obvious and solid, but I'll only post it here after others have commented.
Here you can see overlapped historical bitcoin prices from MtGox with Google search volume for "bitcoin". Conveniently, the "exact" search volume scale (as opposed to the relative scale) mostly matches the USD exchange rate, so it's all on the same unnamed y-axis.
Bitcoin price data are daily average, while google data are unfortunately compounded weekly. I have therefore placed google data points on the 4th day of the week.
Also shown are the same data in log scale so no detail is lost in the valleys.
What do you think? To me, the conclusion is obvious and solid, but I'll only post it here after others have commented.
Jump to: