Portfolio investment decision model based on price forecasting model and risk assessment model

Investors often use gold and Bitcoin to reduce the risk of financial investments by combining them due their low correlation different dynamic correlation. Therefore, it is essential study how a combination two can be invested maximize returns. In this paper, we combine BP neural network model gray analysis from future unpredictability quantification portfolio investment. We want construct price prediction assessment build for investment decisions based on past-day flow only. To specific, paper first builds model. At same time, needs performed in cycle, training set increased each time. Each recorded next day's data aggregated obtain daily prediction. Next, through analysis. order measure riskiness investment, selects three indicators, namely, forecast accuracy, recent trend, volatility. It uses score quantitatively then gives results. Finally, sensitivity was determine model's transaction costs varying bitcoin. found that as become more extensive, number transactions becomes smaller total assets acquired smaller.