A Core Moving Average Indicator based Artificial Market Model for Studying the Bitcoin Trading

Abstract

The Bitcoin cryptocurrency market exhibits complex dynamics that challenge traditional strategy development. The market consists of two types of agents; Random Traders who submit buy or sell orders without any strategic rationale and Chartists trade the finest sets of trading strategies based on the Genetic Algorithms (GAs) theory modelling bitcoin price formation over a representative directive manuscript and recreating BTC price series that displays several formal features found in real-time price series. A key issue of algorithmic trading is the identification parameter configurations that consistently yield profitable outcomes across varying market conditions. To resolve this we simulate trading behavior within an agent-based artificial chartist. The proposed chartists operate using five key indicator including Filter, Exponential Moving Average (EMA), Moving Average Convergence Divergence (MACD), Relative Strength Index (RSI), and Simple Moving Average (SMA). A subset of these agents employs a genetic algorithm to optimize the parameters of these indicators for maximum profitability during a training period, while others use randomly selected parameter configurations. In comparison to a buy and hold place in BTC. Our trading method provide high alpha, utility, sharpe ratio gains and significantly reduce the severity of drawdowns. Simulation results indicate that the genetically optimized chartists using SMA with a range between 7 and 53 achieving a mean value of 21.31 and a standard deviation of 17.13 outperform both randomly parameterized chartists and non-strategic random traders.