Comparison of Active COVID-19 Cases per Population Using Time-Series Models

This research explored the precision of diverse time-series models for COVID-19 epidemic detection in all the thirty-six different states and the Federal Capital Territory (FCT) in Nigeria with the maximum count of daily cumulative of confirmed, recovered and death cases as of 4 November 2020 of COVID-19 and populace of each state. A 14-multi step ahead forecast system for active coronavirus cases was built, analyzed and compared for six (6) different deep learning-stimulated and statistical time-series models using two openly accessible datasets. The results obtained showed that based on RMSE metric, ARIMA model obtained the best values for four of the states (0.002537, 0.001969.12E-058, 5.36E-05 values for Lagos, FCT, Edo and Delta states respectively). While no method is all-encompassing for predicting daily active coronavirus cases for different states in Nigeria, ARIMA model obtains the highest-ranking prediction performance and attained a good position results in other states.

A simple but powerful technique for incorporating a changing underlying inflation trend into standard statistical time series models can improve forecast accuracy significantly—about 20 percent to 30 percent, two to three years out.

The novel coronavirus disease (COVID-19) is regarded as one of the most imminent disease outbreaks which threaten public health on various levels worldwide. Because of the unpredictable outbreak nature and the virus’s pandemic intensity, people are experiencing depression, anxiety, and other strain reactions. The response to prevent and control the new coronavirus pneumonia has reached a crucial point. Therefore, it is essential—for safety and prevention purposes—to promptly predict and forecast the virus outbreak in the course of this troublesome time to have control over its mortality. Recently, deep learning models are playing essential roles in handling time-series data in different applications. This paper presents a comparative study of two deep learning methods to forecast the confirmed cases and death cases of COVID-19. Long short-term memory (LSTM) and gated recurrent unit (GRU) have been applied on time-series data in three countries: Egypt, Saudi Arabia, and Kuwait, from 1/5/2020 to 6/12/2020. The results show that LSTM has achieved the best performance in confirmed cases in the three countries, and GRU has achieved the best performance in death cases in Egypt and Kuwait.

COVID-19 is a disease-causing coronavirus strain that emerged in December 2019 that led to an ongoing global pandemic. The ability to anticipate the pandemic’s path is critical. This is important in order to determine how to combat and track its spread. COVID-19 data is an example of time-series data where several methods can be applied for forecasting. Although various time-series forecasting models are available, it is difficult to draw broad theoretical conclusions regarding their relative merits. This paper presents an empirical evaluation of several time-series models for forecasting COVID-19 cases, recoveries, and deaths in Saudi Arabia. In particular, seven forecasting models were trained using autoregressive integrated moving average, TBATS, exponential smoothing, cubic spline, simple exponential smoothing Holt, and HoltWinters. The models were built using publicly available daily data of COVID-19 during the period of 24 March 2020 to 5 April 2021 reported in Saudi Arabia. The experimental results indicate that the ARIMA model had a smaller prediction error in forecasting confirmed cases, which is consistent with results reported in the literature, while cubic spline showed better predictions for recoveries and deaths. As more data become available, a fluctuation in the forecasting-accuracy metrics was observed, possibly due to abrupt changes in the data.

Gold is known as the most valuable commodity in the world because it is a universal currency recognized by every single bank across the globe. Thus, many people were interested in investing gold since gold market was always steadier compared to other investment (Khamis and Awang, 2020). However, the credibility of gold was questionable due to the changes in gold prices caused by a variety of circumstances (Henriksen, 2018). Hence, information on the inflation of gold prices were needed to understand the trend in order to plan for the future in accordance with international gold price standards. The aim of this study was to identify the trend of Kijang Emas monthly average prices in Malaysia from the year 2010 to 2021, to determine the best fit time series model for Kijang Emas prices in Malaysia and using univariate time series models to forecast Kijang Emas prices in Malaysia. The ARIMA and ARFIMA models were used in this study to model and forecast the prices of gold (Kijang Emas) in Malaysia. Each of the actual monthly Kijang Emas prices for 2021 were found to be within the 95% predicted intervals for both the ARIMA and ARFIMA models. The performances for each model were checked by considering the values of MAE, RMSE and MAPE. From the findings, all the MAE, RMSE and MAPE values showed that the ARFIMA model emerged as the better model in forecasting the Kijang Emas prices in Malaysia compared to the ARIMA model.

In the financial literature, there is great interest in the prediction of stock prices. Stock prediction is necessary for the creation of different investment strategies, both speculative and hedging ones. The application of neural networks has involved a change in the creation of predictive models. In this paper, we analyze the capacity of recurrent neural networks, in particular the long short-term recurrent neural network (LSTM) as opposed to classic time series models such as the Exponential Smooth Time Series (ETS) and the Arima model (ARIMA). These models have been estimated for 284 stocks from the S&P 500 stock market index, comparing the MAE obtained from their predictions. The results obtained confirm a significant reduction in prediction errors when LSTM is applied. These results are consistent with other similar studies applied to stocks included in other stock market indices, as well as other financial assets such as exchange rates.

Nonlinear autoregressive exogenous (NARX), autoregressive integrated moving average (ARIMA) and multi-layer perceptron (MLP) networks have been widely used to predict the appearance value of future points for time series data. However, in recent years, new approaches to predict time series data based on various networks of deep learning have been proposed. In this paper, we tried to predict how various environmental factors with time series information affect the yields of tomatoes by combining a traditional statistical time series model and a deep learning model. In the first half of the proposed model, we used an encoding attention-based long short-term memory (LSTM) network to identify environmental variables that affect the time series data for tomatoes yields. In the second half of the proposed model, we used the ARMA model as a statistical time series analysis model to improve the difference between the actual yields and the predicted yields given by the attention-based LSTM network at the first half of the proposed model. Next, we predicted the yields of tomatoes in the future based on the measured values of environmental variables given during the observed period using a model built by integrating the two models. Finally, the proposed model was applied to determine which environmental factors affect tomato production, and at the same time, an experiment was conducted to investigate how well the yields of tomatoes could be predicted. From the results of the experiments, it was found that the proposed method predicts the response value using exogenous variables more efficiently and better than the existing models. In addition, we found that the environmental factors that greatly affect the yields of tomatoes are internal temperature, internal humidity, and CO2 level.

Abstract BackgroundMany studies have compared the performance of time-series models in predicting pulmonary tuberculosis (PTB). Few studies regarding the role of meteorological factors in predicting PTB are available. This study aims to explore whether incorporating meteorological factors can improve the performance of time series models in predicting pulmonary tuberculosis (PTB).MethodsWe collected the monthly number of PTB cases registered in three cities of China from 2005 to 2018, and data of six meteorological factors in the same period. We constructed three time-series models, including the autoregressive integrated moving average (ARIMA) model, the ARIMA with exogenous variables (ARIMAX) model, and the recurrent neural network (RNN) model. The construction of the ARIMA model did not incorporate meteorological factors, while the construction of ARIMAX and RNN models incorporated meteorological factors. The mean absolute percentage error (MAPE) and root mean square error (RMSE) were used to compare the performance of models in predicting the monthly number of PTB cases in 2018.ResultsBoth the cross-correlation analysis and spearman rank correlation test showed that PTB was related to meteorological factors in the three cities. The prediction performance of both ARIMA and RNN models was improved after incorporating the meteorological factors. The MAPEs of the ARIMA, ARIMAX, and RNN models were 12.536%, 11.957%, and 12.360% in Xuzhou, 15.568%, 11.155%, and 14.087% in Nantong, and 9.700%, 9.660% and 12.501% in Wuxi, respectively. The RMSEs of the three models were 36.194, 33.956 and 34.785 in Xuzhou, 34.073, 25.884 and 31.828 in Nantong, and 19.545, 19.026 and 26.019 in Wuxi, respectively.ConclusionsOur study revealed a possible link between PTB and meteorological factors. Taking meteorological factors into consideration may increase the accuracy of time series models in predicting PTB.

AbstractAn integrated technique combining global climate model (GCM) simulation results and a statistical time series forecasting model (the autoregressive integrated moving average ARIMA model) was developed to bring together the climate change signal from GCMs to city-level historical observations as an approach to obtain location-specific temperature and precipitation projections. This approach assumes that regional temperature and precipitation time series reflect a combination of an underlying climate change signal series and a regional-deviation-from-the-signal series. An ensemble of GCMs is used to describe and provide the climate change signal, and the ARIMA model is used to model and project the regional deviation. Qualitative and quantitative assessments were conducted for evaluating the projection performance of the hybrid GCM-ARIMA (G-ARIMA) model. The results indicate that the G-ARIMA model can provide projected city-specific daily temperature and precipitation series comparable to historical observations and can have improved projection accuracy for several assessed annual indices compared to a commonly used downscaled projection product. The G-ARIMA model is subject to some limitations and uncertainties from the GCM-provided climate change signal. A notable feature of the G-ARIMA model is the efficiency with which projections can be updated when new observations become available, thus facilitating updating of regional temperature and precipitations projections. Given the increasing need for and use of location-specific climate projections in practical engineering applications, the G-ARIMA model is an option for regional temperature and precipitation projection for such applications.