Google search volume can predict price movement of cryptocurrency, suggests study

A working paper circulated by US-based think tank National Bureau of Economic Research suggests that unlike traditional financial markets, the returns and price movements in cryptocurrency markets are a factor of the type of attention they receive.

The paper drafted by Yale University economists Yukun Liu and Aleh Tsyvinski argues that since cryptocurrencies have no exposure to most stock markets and macroeconomic factors, neither have they any exposure to returns of currencies or commodities, a completely different set of factors affect their price movements and thus returns.

Notably, the paper formulates certain parameters which can be used to predict the returns on coins like Bitcoin, XRP and Ethereum.

The paper argues that “high investor attention predicts high future returns over a 1-2 week horizon for Bitcoin, a one-week horizon for Ripple (XRP), and 1-6 week horizon for Ethereum.”

The study, which is based on a financial model called capital asset pricing model or CAPM, states that a one-standard-deviation increase in the Google search for Bitcoin yields a 2.3 percent increase in the next two weeks. Standard deviation is a statistical measure of the spread of data points. The greater the deviation, more spread out the data points (the price in the case of bitcoin).

Similarly, a one-standard-deviation increase in a Twitter posts for bitcoin yields a 2.5 percent increase in the next one week. The paper suggests that negative investor attention adversely affects prices of cryptocurrencies and thus returns. A one-standard-deviation increase in a Google search for 'Bitcoin hack' leads to 2.75 percent decrease in bitcoin returns the following week.

The paper also found that return on Ethereum is exposed to a certain degree to the stock returns of Advanced Micro Devices (AMD), one of the main manufacturers of specialised mining hardware.

“Our main conclusion is that indeed cryptocurrency represents an asset class that can be assessed using simple finance tools. At the same time, cryptocurrencies comprise an asset class which is radically different from traditional asset classes,” Liu and Tsyvinski stated.