Across blockchain networks, governance is a complex task with multiple stakeholders having several financial interests. The process by which decisions are made among these stakeholders is often centralized with certain authorities. For example, investment funds typically are controlled by a central authority. However, this decision-making framework limits openness for global community development and collaboration, as well as the personal autonomy of network participants. It often results in one party controlling the funds for a plethora of members. Furthermore, in a decentralized network, there was no way for investors to safely track financial transaction data and have it stored remotely, enabling global access.
One particular challenge we faced was differentiating our solution from other DOAs that were in the financial sector. We realized that by making Algorand the basis of all of our transactions and investments, we were introducing potential volatility. To help reduce volatility, we wanted to know the optimal time to invest in our portfolios. So, we decided to make a neural network that could make price predictions for not only Algo but also for other cryptocurrencies and financial securities.
To start, we aggregated a dataset with financial information for top cryptocurrency chains and hardware technology companies. We then ran statistical analysis on public Algorand data using the Python Pandas Library. Next, we decided to expand the dataset. The dataset now includes public financial data for Apple, Intel, NVIDIA, Bitcoin, Ethereum, and Algorand. We then began construction on a neural network to make daily price predictions.