(An) empirical study on the existence of the psychological barrier and its change over time in bitcoin markets

Abstract

In this paper, we examine the existence of psychological barriers in three Bitcoin markets, Coinbase of the US, Bitstamp of Europe and Bitfinex of Hong Kong. Using barrier proximity and hump tests, we found consistent evidence for the simultaneous existence of price barriers and clustering

barriers at the 1000- and 10000-level, and clustering at the 10- and 100-level. Also, we examine, for the first time, the time-varying trends of the price barriers in the three markets by using the coefficients from the proximity tests. The overall trend of barriers seems to drop at the 1000-level barrier after the historically high price, but the 10000-level barrier appears at the same moment with sufficient magnitudes, indicating a psychological barrier "jumping". This paper contributes to the literature on psychological barriers and price clustering, and also on market efficiency and market anomalies.