Cryptocurrency Price Estimation Using Hyperparameterized Oscillatory Activation Functions in LSTM Networks

Activation functions are critical components of neural networks, helping the model learn highly-intricate dependencies, trends, and patterns. Non-linear activation functions allow the model to behave as a functional approximator, learning complex decision boundaries and multi-dimensional patterns in the data. Activation functions can be combined with one another to learn better representations with the objective of improving gradient flow, performance metrics reducing training time and computational cost. Recent work on oscillatory activation functions\cite{noel2021growing}\cite{noel2021biologically} showcased their ability to perform competitively on image classification tasks using a compact architecture. Our work proposes the utilization of these oscillatory activation functions for predicting the volume-weighted average of Bitcoin on the G-Research Cryptocurrency Dataset. We utilize a popular LSTM architecture for this task achieving competitive results when compared to popular activation functions formally used.

Followness of Altcoins in the Dominance of Bitcoin: A Phase Analysis

Due to the transparency, simplicity, and blockchain system, cryptocurrencies gained popularity in the modern world. This led to more use of cryptocurrencies for speculation and investment rather than a medium of exchange. It is crucial to analyse the nature of the crypto market before investing in such currencies. With this intention, the paper tried to know the extent of following (Followness) of altcoins to the bitcoin in the different dominance phases like High Dominance, Low Dominance, and Moderate Dominance. For this purpose, daily closing prices of the Bitcoin and five major altcoins (Ethereum, Litecoin, Namecoin, Doge, and Ripple) are collected for the last five years and analyse the relationship between bitcoin and altcoins. Pearson's correlation coefficient test is used to know the direction of the relationship, and Vector Error Correction Model is used to see the extent of the relation. In general, the empirical result of the study showed cointegration between bitcoin and Altcoin. It also depicted that Altcoin showed a high level of followness in the moderate dominance phase and low followness in the low dominance phase. The study developed a price estimation equation to predict the price of altcoins depending upon the price of bitcoin and its dominance in the crypto market. This paper concludes that the dominance of Bitcoin also has a significant role in the price movement of altcoins.

Cal-Maine Foods, Inc.: stock price estimation in the midst of pandemic

This case concerns an analyst’s task to value Cal-Maine Foods, Inc., the largest and only publicly traded U.S. egg production firm. The case takes place in 2020, at the time of the Covid-19 pandemic. Historically volatile egg prices were even more volatile in April 2020, with a large spike in prices that led the state of Texas to sue the firm for price gouging. Added to this, Cal-Maine had an unexpectedly bad earnings report a few months earlier, and prior to that, the firm cut its dividend. How should the analyst incorporate these shocks – or should they be included at all? How can the analyst assess the risk of a company that has volatile revenues and costs and a widely varying beta? Which factors is the analysis most sensitive to? Was the market overvaluing Cal-Maine? Or, was there potential for investors to profit from investing in the firm?