In India, any crypto enthusiast who has done the math knows that mining such currencies is not profitable; It makes more sense to guess instead. Unfortunately, since cryptocurrencies are not tied to products and services, but are largely driven by investor sentiment, traditional methods of financial speculation do not apply.
Indraprastha Institute of Information Technology Delhi (IIIT-Delhi) researchers have developed an artificial intelligence (AI) driven approach to predicting cryptocurrency prices, as per an official statement.
The work of Ph.D. scholar of IIIT-Delhi Shalini Sharma and her supervisor Dr. Angshul Majumdar is an artificial intelligence (AI)-driven approach to predicting cryptocurrency prices.
According to the statement, Shalini Sharma, a PhD scholar of IIIT-Delhi and Angshul Majumdar, Sharm’s supervisor started with the conventional Baum-Welch framework but incorporated deep learning within it.
Furthermore, it mentioned that there are two predominant approaches for financial (or any) prediction. One is based on conventional methods of the 70s, the celebrated Baum-Welch approach to making money in stock markets.
“The information about uncertainty is crucial in making financial decisions. However, this approach requires knowledge about the underlying events causing the changes in price; this information is not available for cryptocurrencies,” it said.
The work has recently been accepted in Elsevier Informatics. Based on available cryptocurrency data, it was shown that the said method is “very accurate” in predicting future prices. It can beat all state-of-the-art methods in terms of accuracy. In addition, the uncertainty estimates make the results more interpretable, the institute claims.
The other approach, according to it, is modern, driven by a bleeding-edge AI called deep learning. “Deep learning is abstract but does not require knowledge/information about underlying factors. Unfortunately, it cannot give uncertainties about the predictions; this precludes the interpretability aspects,” the statement said.
Researchers at IIIT-Delhi started with the traditional Baum-Welch framework, but incorporated deep learning within it. This incorporation enabled him to make predictions without knowledge of the underlying events.
Since the technique was developed on the basis of Baum-Welch, it makes interpretable predictions and also gives an estimate of the uncertainty surrounding the prediction.
There are two predominant approaches for financial (or any) prediction. One is based on conventional methods of the 70s, the celebrated Baum-Welch approach to making money in stock markets. It is interpretable and gives us not only the predicted values but also the uncertainty about the prediction.
The information about uncertainty is crucial in making financial decisions. However, this approach requires knowledge about the underlying events causing the changes in price; this information is not available for cryptocurrencies.
The researchers showed that the uncertainty estimates obtained from their method are correlated with historical crypto volatility index (CVI) values, which refer to how much a cryptocurrency fluctuates with time. As a result, the method aimed to prove that the predictions made by their method are interpretable.
Dr. Majumdar says, “it is very important to have accurate yet interpretable predictions in these cases,otherwise one won’t have confidence in the approach”.