![]() Specifically, we show that these technical indicators do indeed provide accurate predictions of the COVID-19 pandemic, in the sense that they are statistically significant. We propose here that these techniques, collectively known as “technical analysis”, developed to give buy/sell signals in the stock market (in particular ‘momentum’ indicators) can be used to provide predictions for other time series data, such as COVID-19 forecasts. Drawing on this connection we hypothesized that strategies developed for predicting stock price movements can be repurposed to forecast changes in the trend of the number of new COVID cases, and more generally any system that is well described as a non-stationary random walk. On the other hand, the daily increase in COVID-19 cases can be modeled as a random walk due to the complex nature of human interactions and it has an overall trend as an infectious disease in the spread and controlled phases. ![]() In the context of stocks, the random walk hypothesis can be formulated with the daily rate of returns in the stock market randomly drawn from a Gaussian or Laplace distribution 1. While fluctuations in the number of new COVID cases and the prices of stocks may naively seem disconnected, both systems can be described as non-stationary random walks, i.e. a time series which exhibits random fluctuations around a longer-term trend. Finally, our new methods are used to assess whether current health policies are impacting the growth in new COVID-19 cases. In particular, we show the utility of this novel approach by considering the identification of the beginnings of subsequent waves of the pandemic. By providing a quantitative assessment of MACD, RSI, and candlestick analyses, we show their statistical significance in making predictions for both stock market data and WHO COVID-19 data. By reinterpreting COVID-19 daily cases in terms of candlesticks, we are able to apply some of the most popular stock market technical indicators to obtain predictive power over the course of the pandemics. Notably, reliable short term forecasting can provide potentially lifesaving insights into logistical planning, and in particular, into the optimal allocation of resources such as hospital staff and equipment. We propose that technical analysis tools developed to give buy/sell signals in asset trading can be applied to analyze time series datasets in the natural sciences, and we show this explicitly for a study of WHO COVID-19 data.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |