Research between investor sentiment and bitcoin price based on the VAR model

Abstract

In this paper, we examine the dynamic relationship between Bitcoin prices and investor sentiment indicators. According to the correlation, the five time series variables of S&P cryptocurrency extensive digital market index, S&P 500 value index, bitcoin trading volume and the original Baidu index were reasonably selected to establish the comprehensive index of investor sentiment. The vector autoregressive VAR model is used to verify the relationship between the Bitcoin price (CLOSE) and the sentiment indicators. According to the Granger causality test and the pulse response function analysis, there is a bidirectional causal relationship between the two. The constructed indicators have important practical significance.