Time-frequency connectedness among clean-energy stocks and fossil fuel markets: Comparison between financial, oil and pandemic crisis

Nonlinear dependence and connectedness between clean/renewable energy sector equity and European emission allowance prices

Are clean energy assets a safe haven for international equity markets

News-based equity market uncertainty and crude oil volatility

Crude oil prices and clean energy stock indices: Lagged and asymmetric effects with quantile regression

Unlike previous studies examining the association between crude oil and renewable energy stock prices under average conditions, we employ a quantile-based regression approach offering a more comprehensive dependence structure under diverse market conditions. Using weekly data covering crude oil prices (WTI market) and three clean energy stock indices (the Wilderhill Energy Index, MAC Global Solar Energy Index, and S&P Global Clean Energy Index), quantile regression analyses provide solid evidence of the decreasing dependence of clean energy stock returns on crude oil returns. The lagged effect of WTI oil returns on clean energy stock returns is generally significant, which indicates that clean energy stock returns react differently to new information on oil returns under different market conditions. We further check for asymmetrical effects of oil returns on clean energy stock returns in various market conditions and find a strong effect of negative oil returns during bearish periods and an insignificant effect during bullish episodes.

Crude oil prices and clean energy stock indices : Lagged and asymmetric effects with quantile regression

The COVID-19 pandemic has put the entire world in an unprecedented situation., claiming thousands of lives. COVID-19 is now present in over 212 countries and territories., where it poses a significant risk to the public health with the rising incidence of infection cases and significant death tolls. The RT -PCR(Reverse transcription-polymerase chain reaction) test is the most popular diagnostic tool for coronavirus detection. This test was time-consuming and also required additional supplies and labor. Additionally., many nations initially lacked RTPCR kits., and the unproven accuracy of these kits cleared the path for the use of X-ray pictures coupled with deep-learning techniques for more conclusive analysis. This research work proposes a covid-19 detection system using deep learning models to detect covid-19 virus from other disease images. The work comprises a detailed review of previous research and proposes methods with different algorithms to detect Covid-19. Four different models used in this paper are VGGNet., LeNet5., AlexNet., and a custom on a publicly available repository.

Deep learning based Detection of Coronavirus (COVID-19) using Chest X-ray images

Human movement has a significant influence on disease propagation. To suppress an outbreak, movement restrictions are imposed in locations classified as disease hotspots. It is essential to monitor the sickness on a local level in order to identify locations that need concentrated attention in order for several nations to succeed in preventing the further spread of COVID-19 without reimposing national bans. The goal of this research is to identify epidemic disease hotspots within local communities by utilizing publicly available health- related data and unstructured data analysis methods. We used data that was self-reported from the whole population and obtained via a mobile application, in combination with data from targeted PCR(polymerase chain reaction) testing, to develop estimates of the frequency and incidence of sickness across the area. We also demonstrate how these estimations may be utilized to identify infection hotspots in real time. A regression analysis is done based on the numerical data to determine the best regression model for these important parameters. The proposed regression models are useful for estimating hotspots under various ventilation settings. We used the ARIMA(Autoregressive Integrated Moving Average) and SARIMAX(Seasonal Auto-Regressive Integrated Moving Average with exogenous factors) models to identify hotspots for future COVID waves using a data set of COVID-19 cases.

Prediction of epidemic disease cases using ARIMA and SARIMAX models

Credit card fraud has increased due to the growing usage of credit cards for in-store and online transactions. These cards allow the user to make substantial purchases without carrying much money. They have simplified payment for clients and ushered in a new era of cashless commerce. Despite its many advantages, this kind of digital currency transaction is not without its own special risks. The incidence of credit card fraud increases proportionally with the growth of the user base. Therefore, it is essential to create models and solutions that prevent a user or community from falling victim and suffering financial and reputational losses. Typically, these methods are employed alone or in combination with ensemble or conceptual approaches to generate classifiers. This paper gives an overview of the fraud detection system using various supervised Machine Learning approaches i.e., Extreme Gradient Boosting (XGB), Decision Tree (DT), Logistic Regression (LR) and, Random Forest (RF), and compares their performance on factors like accuracy, sensitivity, etc. on a dataset for European cardholders obtained from Kaggle.

Ishaan Dawar

Papers

Crude oil prices and clean energy stock indices: Lagged and asymmetric effects with quantile regression

Crude oil prices and clean energy stock indices : Lagged and asymmetric effects with quantile regression

Deep learning based Detection of Coronavirus (COVID-19) using Chest X-ray images

Prediction of epidemic disease cases using ARIMA and SARIMAX models

Supervised Learning Methods for Identifying Credit Card Fraud