Gold and Bitcoin Price Forecasting Based on Multiple Models
- Select Volume / Issue:
- Type of Publication:
- Quantitative Trading, Time Series Analysis, Forecasting Model, ARIMA Model
- Zimu Lin; Han Wang; Xiang Zheng; Rongrong Shentu; Luyan Wang
- Gold and Bitcoin are volatile assets traded in two markets. For centuries, people have been using gold as a wealth preservation tool. With the increasing role of the gold market, statistical modeling and forecasting of gold market fluctuations and prices have attracted widespread attention. With the development of data, modern currency presents the characteristics of virtualization, and some data represented by Bitcoin have achieved rapid development. However, due to the influence of policy, economic situation and many uncertain factors, the transaction price of Bitcoin shows the characteristics of non-linearity, non-stationarity and high volatility, which puts forward a great problem for the prediction of Bitcoin price. In this regard, we introduce the ARIMA model to predict the prices of gold and bitcoin in the next three days while considering the influence of time factors, and select some other prediction models to compare with the performance of the ARIMA model to explore our time series prediction method is good, which also shows that the model can be used as a reference investment model for traders.
Full text: IJISM_983_FINAL.pdf [Bibtex]