Dynamical System Modeling and Stability Investigation

DSMSI-2023

Dedicated to the 77th anniversary of the outstanding Ukrainian scientist

professor Denys Khusainov

December 19-21, 2023, Kyiv, Ukraine

Dynamic Rebalancing of Cryptocurrency Portfolio Based on Forecasted Technical Indicators and Random Forest Method

​

Olena Liashenko

Tetyana Kravets

Vadym Proshchenko

Introduction

Since the introduction of Bitcoin in 2009, cryptocurrency has attracted a lot of interest among investors due to its dynamic nature and great potential for high returns. At the same time, its volatility provides both the opportunity for large gains and can lead to the loss of a significant amount of capital. Many portfolio optimization strategies have been proposed, which are usually based on traditional investment approaches, such as the Markowitz portfolio theory. However, given the peculiarities of cryptoassets, there is a need for new, adapted approaches to the optimization of cryptocurrency portfolios. One such approach is dynamic portfolio optimization based on machine learning methods. This approach allows for predicting future trends in the cryptocurrency market and adjusting the portfolio to these forecasts.

Dynamical System Modeling and Stability Investigation, DSMSI-2023

Purpose of study

Build a model for rebalancing the portfolio of cryptoassets based on the forecasted indicators of technical analysis using the Random Forest method.

Dynamical System Modeling and Stability Investigation, DSMSI-2023

Technical indicators are mathematical calculations that use price and/or volume to predict future price movements.

• A moving average determines the average price of an asset over a certain period and is used to detect trends. The exponential moving average (EMA) is calculated by the formula:

​

​

​

• Moving Average Convergence/Divergence (MACD) is calculated as the difference between the short-term and long-term EMA. It is a trend-following tool that uses moving averages to determine the momentum.
• The signal line is defined as the EMA of the MACD line. Combined analysis of the signal line with the MACD line can be useful for identifying potential reversals or entry and exit points.
• Histogram (MACDH) gives a graphical representation of the divergence and convergence of the MACD line and the signal line.

​

Technical Analysis in Crypto Trading

Dynamical System Modeling and Stability Investigation, DSMSI-2023

• MACDH is sensitive to "sawtooth effects" - minor fluctuations in the price lead to frequent and significant fluctuations in the value of the indicator. To solve the problem of minor fluctuations near the zero axis, a modification of this index was introduced:

​

​

​

​

• Technical analysis tools are best used in combination with each other, along with fundamental analysis or machine learning elements. One of the methods of such synthesis is the forecasted MACDH% (fMACDH%), which is based on the so-called “forecasted” analogs of the EMA and MACD indicators:

Technical Analysis in Crypto Trading

Dynamical System Modeling and Stability Investigation, DSMSI-2023

Proportional dynamic rebalancing of the portfolio according to the Alpha-Beta fMACDH% criterion

Dynamical System Modeling and Stability Investigation, DSMSI-2023

Block diagram of one iteration of the algorithm

The 10 cryptocurrencies with the largest market capitalization as of October 10, 2022, were selected for the study:

• Bitcoin;
• Ethereum;
• Tether;
• USD Coin;
• BNB;
• XRP;
• Binance USD;
• Cardano;
• Solana;
• Dogecoin.

Prices were taken in the period from September 9, 2020, to October 9, 2022, that is, 761 observations.

Assets selection

Dynamical System Modeling and Stability Investigation, DSMSI-2023

The ratio between the mean value and the standard deviation of R2 of Random Forest models build based on different number of accounted periods

Random Forest forecasting method

Dynamical System Modeling and Stability Investigation, DSMSI-2023

Used modified algorithm

Dynamical System Modeling and Stability Investigation, DSMSI-2023

Change in portfolio value for a model with parameters Alpha=Beta=0, w=0.5

Dynamical System Modeling and Stability Investigation, DSMSI-2023

Change in portfolio value depending on the Alpha and Beta parameters

Dynamical System Modeling and Stability Investigation, DSMSI-2023

Depending on the Alpha

Depending on the Beta

Change in portfolio value for a model with parameters Alpha=2, Beta=-0.5, w=0.5

Dynamical System Modeling and Stability Investigation, DSMSI-2023

Key conclusions from the evaluation of the model

Dynamical System Modeling and Stability Investigation, DSMSI-2023

• The model can achieve high ROI.
• The portfolio showed volatility with a mix of growth, decline, and stability, necessitating further research into the contributing factors.
• Adjusting the Alpha, Beta, and w parameters significantly affected portfolio performance. Higher Alpha values indicated smoother declines and increases, suggesting a preference for lower-risk investors.
• Different combinations of Alpha and Beta values resulted in varying levels of portfolio volatility and diversification.
• Higher Alpha and Beta values tended to lower portfolio volatility but also decreased diversification, suggesting a trade-off between risk and reward.
• Incorporating a predictive component (w parameter) enhanced model effectiveness and reduced the impact of minor fluctuations.

Discussion

Cryptocurrencies are attractive for portfolio diversification due to high volatility and potential profits.

The study confirms that dynamic rebalancing using the fMACDH% indicator and Random Forest method effectively increases portfolio value. Adjusting Alpha and Beta impacts portfolio volatility and value, higher values reduce volatility but may limit gains. The parameter w is crucial for trade frequency and efficiency.

Future research should explore other prediction methods, optimize parameters, and integrate macroeconomic and sentiment analysis data to enhance the model.

Dynamical System Modeling and Stability Investigation, DSMSI-2023

Thank you for your attention