scholarly journals MODELLING OF COVID-19 OUTBREAK INDICATORS IN CHINA BETWEEN JANUARY AND APRIL

2020 ◽  
Author(s):  
Senol Çelik ◽  
Handan Ankarali ◽  
Ozge Pasin
Keyword(s):  
Time Series ◽  
Nonlinear Models ◽  
Linear Trend ◽  
Growth Curves ◽  
Information Criteria ◽  
Suitable Model ◽  
Times Series ◽  
Time Series Method ◽  
The Times ◽  
Negative Exponential

AbstractBackgroundThe aim of this study is to explain the changes of outbreak indicators for coronavirus in China with nonlinear models and time series analysis. There are lots of methods for modelling. But we want to determine the best mathematical model and the best time series method among other models.MethodsThe data was obtained between January 22 and April 21, 2020 from China records. The number of total cases and the number of total deaths were used for the calculations. For modelling Weibull, Negative Exponential, Von Bertalanffy, Janoscheck, Lundqvist-Korf and Sloboda models were used and AR, MA, ARMA, Holt, Brown and Damped models were used for time series. The determination coefficient (R2), Pseudo R2 and mean square error were used for nonlinear modelling as criteria for determining the model that best describes the number of cases, the number of total deaths and BIC (Bayesian Information Criteria) was used for time series.ResultsAccording to our results, the Sloboda model among the growth curves and ARIMA (0,2,1) model among the times series models were most suitable models for modelling of the number of total cases. In addition Lundqvist-Korf model among the growth curves and Holt linear trend exponential smoothing model among the times series models were most suitable model for modelling of the number of total deaths. Our time series models forecast that the number of total cases will 83311 on 5 May and the number of total deaths will be 5273.ConclusionsBecause results of the modelling has providing information on measures to be taken and giving prior information for subsequent similar situations, it is of great importance modeling outbreak indicators for each country separately.

scholarly journals Modeling of COVID-19 Outbreak Indicators in China Between January and June

2020 ◽  
pp. 1-9
Author(s):  
Senol Celik ◽  
Handan Ankarali ◽  
Ozge Pasin
Keyword(s):  
Time Series ◽  
Mean Square Error ◽  
Bayesian Information Criterion ◽  
Linear Trend ◽  
Growth Curves ◽  
Information Criterion ◽  
Time Series Models ◽  
Cumulative Number ◽  
Mean Square ◽  
Nonlinear Growth

ABSTRACT Objectives: The objective of this study is to compare the various nonlinear and time series models in describing the course of the coronavirus disease 2019 (COVID-19) outbreak in China. To this aim, we focus on 2 indicators: the number of total cases diagnosed with the disease, and the death toll. Methods: The data used for this study are based on the reports of China between January 22 and June 18, 2020. We used nonlinear growth curves and some time series models for prediction of the number of total cases and total deaths. The determination coefficient (R2), mean square error (MSE), and Bayesian Information Criterion (BIC) were used to select the best model. Results: Our results show that while the Sloboda and ARIMA (0,2,1) models are the most convenient models that elucidate the cumulative number of cases; the Lundqvist-Korf model and Holt linear trend exponential smoothing model are the most suitable models for analyzing the cumulative number of deaths. Our time series models forecast that on 19 July, the number of total cases and total deaths will be 85,589 and 4639, respectively. Conclusion: The results of this study will be of great importance when it comes to modeling outbreak indicators for other countries. This information will enable governments to implement suitable measures for subsequent similar situations.

scholarly journals Spatial and Temporal Spread of the Coronavirus Pandemic using Self Organizing Neural Networks and a Fuzzy Fractal Approach

2021 ◽  
Author(s):  
Patricia Melin ◽  
Oscar Castillo
Keyword(s):  
Neural Networks ◽  
Time Series ◽  
Temporal Trends ◽  
Temporal Analysis ◽  
Virus Propagation ◽  
Self Organizing Maps ◽  
Fractal Approach ◽  
Times Series ◽  
The Times ◽  
Self Organizing

In this article, the evolution in space and in time of the coronavirus pandemic is studied by utilizing a neural network with a self-organizing nature for the spatial analysis of data, and a fuzzy fractal method for capturing the temporal trends of the time series of the countries. Self-organizing neural networks possess the capability for clustering countries in the space domain based on their similar characteristics with respect to their coronavirus cases. In this form enabling finding the countries that are having similar behavior and thus can benefit from utilizing the same methods in fighting the virus propagation. To validate the approach, publicly available datasets of coronavirus cases worldwide have been used. In addition, a fuzzy fractal approach is utilized for the temporal analysis of time series of the countries. Then, a hybrid combination of both the self-organizing maps and the fuzzy fractal approach is proposed for efficient COVID-19 forecasting of the countries. Relevant conclusions have emerged from this study, that may be of great help in putting forward the best possible strategies in fighting the virus pandemic. A lot of the existing works concerned with the Coronavirus have look at the problem mostly from the temporal viewpoint that is of course relevant, but we strongly believe that the combination of both aspects of the problem is relevant to improve the forecasting ability. The most relevant contribution of this article is the proposal of combining neural networks with a self-organizing nature for clustering countries with high similarity and the fuzzy fractal approach for being able to forecast the times series and help in planning control actions for the Coronavirus pandemic.

scholarly journals THE RETAIL AND DISTRIBUTION GASOLINE MARKET IN BRAZIL - A STUDY OF VARIATION PRICE BEHAVIORS

2018 ◽  
Vol 3 (2) ◽  
Author(s):  
Aloisio S. N. Filho ◽  
Thiago Barros Murari ◽  
Marcelo A. Moret
Keyword(s):  
Time Series ◽  
Gasoline Prices ◽  
Computational Techniques ◽  
Times Series ◽  
Conventional Analysis ◽  
Exogenous Factors ◽  
Type C ◽  
Cumulative Sum Control ◽  
The Times ◽  
Economic Regions

In this paper evaluates the effects in the gasoline prices after the Brazilian downstream oil chain liberation, in late 1990s. That stage meant that the Brazilian govern, that no longer setting the maximum and minimum values of all fuels. For this purpose, the gasoline type C prices were collected from fifteen relevant cities in five economic regions of Brazil, between the years 2005 and 2014. The sequences of computational techniques were applied on these datasets. The stationary and linearity for variation prices time series were analyzed in all cities and, also, the correlations among all cities in order to recognize the times series patterns. Furthermore, the Cumulative Sum control (CUMSUM) chart was used to detect smaller parameter shifts on the distribution time series. Our results reveled distinct patterns for middle of 2005 and the middle of 2006, and also for the first months of 2011 and the middle of 2012. Reinforcing the idea of the Brazilian retail and distribution are governed strongly by exogenous factors. This makes a conventional analysis difficult to be used. Once, the Brazilian downstream fuel chain suggests to be a complexity system.

scholarly journals Forecasting Randomly Distributed Zero-Inflated Time Series

2017 ◽  
Vol 17 (1) ◽  
pp. 7-19
Author(s):  
Mariusz Doszyń
Keyword(s):  
Time Series ◽  
Forecast Error ◽  
Real Data ◽  
Economic Time Series ◽  
Times Series ◽  
Ex Post ◽  
Data Regression ◽  
Economic Time ◽  
The Times ◽  
A Company

Abstract The main aim of the article is to propose a forecasting procedure that could be useful in the case of randomly distributed zero-inflated time series. Many economic time series are randomly distributed, so it is not possible to estimate any kind of statistical or econometric models such as, for example, count data regression models. This is why in the article a new forecasting procedure based on the stochastic simulation is proposed. Before it is used, the randomness of the times series should be considered. The hypothesis stating the randomness of the times series with regard to both sales sequences or sales levels is verified. Moreover, in the article the ex post forecast error that could be computed also for a zero-inflated time series is proposed. All of the above mentioned parts were invented by the author. In the empirical example, the described procedure was applied to forecast the sales of products in a company located in the vicinity of Szczecin (Poland), so real data were analysed. The accuracy of the forecast was verified as well.

scholarly journals Modeling nonlinearities with mixtures-of-experts of time series models

2006 ◽  
Vol 2006 ◽  
pp. 1-22 ◽  
Author(s):  
Alexandre X. Carvalho ◽  
Martin A. Tanner
Keyword(s):  
Time Series ◽  
Latent Variable ◽  
Nonlinear Models ◽  
Simulated Data ◽  
Likelihood Estimation ◽  
Information Criteria ◽  
Time Series Models ◽  
Family Time ◽  
Mixtures Of Experts ◽  
Standard Information

We discuss a class of nonlinear models based on mixtures-of-experts of regressions of exponential family time series models, where the covariates include functions of lags of the dependent variable as well as external covariates. The discussion covers results on model identifiability, stochastic stability, parameter estimation via maximum likelihood estimation, and model selection via standard information criteria. Applications using real and simulated data are presented to illustrate how mixtures-of-experts of time series models can be employed both for data description, where the usual mixture structure based on an unobserved latent variable may be particularly important, as well as for prediction, where only the mixtures-of-experts flexibility matters.

Institutions and growth: the times-series and cross-section evidence

2011 ◽  
Vol 7 (4) ◽  
pp. 567-570 ◽  
Author(s):  
JAIME ROS
Keyword(s):  
Economic Growth ◽  
Time Series ◽  
Cross Section ◽  
Comprehensive Analysis ◽  
Development Theory ◽  
Times Series ◽  
The Times ◽  
Economic Development Theory ◽  
The Relationship ◽  
Made In

Abstract:In his comprehensive analysis of the relationship between institutions and economic growth, Ha-Joon Chang, in his article ‘Institutions and Economic Development: Theory, Policy and History’, reviews the empirical evidence on this relationship emphasizing the contrast between the conclusions that one can derive from the time-series evidence and the claims often made in favor of ‘liberalized institutions’ based on the results of cross-section studies. Does the time-series evidence contradict the results of cross-section studies regarding the relationship between institutions and growth? In this comment, I argue that in stressing the contrast between these two kinds of evidence, Chang falls short of a full criticism, consistent with his theoretical analysis, of cross-section studies while at the same time failing to infer what the time-series evidence really shows.

scholarly journals ANALYSIS OF REGIONAL INCOME DEVELOPMENT (Analisis Perkembangan Pendapatan Asli Daerah (PAD))

2018 ◽  
Vol 1 (2) ◽  
Author(s):  
Barmin Yusuf
Keyword(s):  
Linear Trend ◽  
Regional Government ◽  
Analysis Model ◽  
Regional Income ◽  
Local Revenue ◽  
Times Series ◽  
Wealth Management ◽  
Income Development ◽  
The Government ◽  
The Times

Abstract The development budget needed every year is increasing. For this reason the regional government must improve regional economic development with a targeted level of GDP to be proportional to the amount of regional expenditure. The government also determines the sources of revenue from regions that can be extracted. One of the most important sources of income is original income (PAD). The original regional income consists of regional taxes and levies, profits of regionally owned companies, results of regional wealth management, and other local revenue. Regional tax is a source of local revenue which has a significant role in the formation of local revenue. The purpose of this study was to determine the development of Regional Original Revenue (PAD) in North Gorontalo District. The quantitative analysis model is used to analyze the development of North Gorontalo Regency's Local Revenue (PAD) using the Times Series or the time needed to look at revenue potential, namely the Linear Trend method. In the results of this study stated that the original revenue (PAD) of North Gorontalo Regency in the future will continue to increase. Where it can be seen that in 2015 the original regional income (PAD) of North Gorontalo District which was compared to 2014 would have decreased by 983,042,804 Rupiah and in 2016 increased by 1,733,651,583 Rupiah when compared to 2014, while in 2017 North Gorontalo regency's local revenue (PAD) when compared to 2014 increased by 4,450,345,972 Rupiah. Abstrak Anggaran pembangunan yang dibutuhkan tiap tahun semakin meningkat. Untuk itu pemerintah daerah harus meningkatkan perkembangan ekonomi daerah dengan tingkat PDB yang ditargetkan agar sebanding dengan jumlah pengeluaran daerah. Pemerintah pun menentukan sumber-sumber penerimaan daerah yang dapat digali. Salah satu sumber penerimaan paling penting adalah pendapatan asli daerh (PAD).Pendapatan asli daerah terdiri dari pajak dan retribusi daerah, keuntungan perusahaan milik daerah, hasil pengelolaan kekayaan daerah, dan lain-lain pendapatan asli daerah. Pajak daerah merupakan sumber pendapatan asli daerah yang cukup besar peranannya dalam terbentuknya pendapatan asli daerah.Tujuan dari penelitian ini adalah untuk mengetahui perkembangan Pendapatan Asli Daerah (PAD)  di Kabupaten Gorontalo Utara. Model analisis kuantitatif digunakan untuk menganalisa perkembangan Pendapatan Asli Daerah (PAD) Kabupaten Gorontalo Utara menggunakan Times Series atau runtut waktu untuk melihat potensi pendapatan yaitu metode Trend Linier.  Dalam hasil penelitian ini menyatakan pendapatan asli daerah (PAD) Kabupaten Gorontalo Utara dimasa yang akan datang akan terus mengalami peningkatan. Dimana dapat dilihat bahwa pada tahun 2015 pendapatan asli daerah (PAD) Kabupaten Gorontalo Utara yang ada apabila dibandingkan dengan tahun 2014 akan mengalami penurunan sebesar 983.042.804 Rupiah dan pada tahun 2016 meningkat sebesar 1.733.651.583 Rupiah apabila dibandingkan dengan tahun 2014, sedangkan tahun 2017 pendapatan asli daerah (PAD) Kabupaten Gorontalo Utara apabila dibandingkan dengan tahun 2014 meningkat sebesar 4.450.345.972 Rupiah. 

scholarly journals Research of Forecasting on Tourist Arrivals to Malaysia

2019 ◽  
Vol 8 (12S2) ◽  
pp. 686-689
Keyword(s):  
Time Series ◽  
Time Series Data ◽  
Linear Trend ◽  
Study Data ◽  
Series Data ◽  
Data Series ◽  
Suitable Model ◽  
Absolute Deviation ◽  
Double Exponential Smoothing ◽  
High Level

Tourists get attracted towards Malaysia because of our culture and geography. Apart from heritage and culture, the tourists from all over the world visit here for various purpose. Therefore, forecasting tourist arrivals with high level of accuracy becomes important because it can ensure the development of tourism industries. So, this study focuses on tourist arrivals to Malaysia. This paper attempts to define the component of patterns exist in the time series data, to determine the most suitable model best fits in data series by using the error measure that are Mean Square Error (MSE) and Mean Absolute Deviation (MAD) and to forecast the one-step ahead forecast on the best model. In this study, data of tourist arrivals to Malaysia has been obtained from January 2000 until December 2018. All 228 monthly data were analyzed by using selected Univariate Modeling. The result found that tourist arrivals to Malaysia has a linear trend model and Double Exponential Smoothing with α = 0.17 was the best model for this time series.

scholarly journals Development of Algorithms for Choosing the Best Time Series Models and Neural Networks to Predict COVID-19 Cases

2021 ◽  
Vol 21 (3) ◽  
pp. 26-35
Author(s):  
M.S.A. Abotaleb ◽  
◽  
T.A. Makarovskikh
Keyword(s):  
Neural Network ◽  
Time Series ◽  
Network Model ◽  
Neural Network Model ◽  
Nonlinear Models ◽  
Linear Trend ◽  
Selection Criterion ◽  
Network Models ◽  
Neural Network Models ◽  
Chelyabinsk Region

Time series analysis became one of the most investigated fields of knowledge during spreading of the COVID-19 around the world. The problem of modeling and forecasting infection cases of COVID-19, deaths, recoveries and other parameters is still urgent. Purpose of the study. Our article is devoted to investigation of classical statistical and neural network models that can be used for forecasting COVID-19 cases. Materials and methods. We discuss neural network model NNAR, compare it with linear and nonlinear models (BATS, TBATS, Holt's linear trend, ARIMA, classical epidemiological SIR model). In our article we discuss the Epemedic.Network algorithm using the R programming language. This algorithm takes the time series as input data and chooses the best model from SIR, statistical models and neural network model. The model selection criterion is the MAPE error. We consider the implementation of our algorithm for analysis of time series for COVID -19 spreading in Chelyabinsk region, and predicting the possible peak of the third wave using three possible scenarios. We mention that the considered algorithm can work for any time se-ries, not only for epidemiological ones. Results. The developed algorithm helped to identify the pat-tern of COVID -19 infection for Chelyabinsk region using the models realized as parts of the consi-dered algorithm. It should be noted that the considered models make it possible to form short-term forecasts with sufficient accuracy. We show that the increase in the number of neurons led to in-creasing accuracy, as there are other cases where the error is reduced in case of reducing the number of neurons, and this depends on COVID -19 infection spreading pattern. Conclusion. Hence, to get a very accurate forecast, we recommend re-running the algorithm weekly. For medium-range fore-casting, only the NNAR model can be used from among those considered but it also allows to get good forecasts only with horizon 1–2 weeks.

Time Series Analysis and Mining Techniques

2011 ◽  
pp. 1120-1124
Author(s):  
Mehmet Sayal
Keyword(s):  
Time Series ◽  
Time Series Analysis ◽  
Time Series Data ◽  
Series Data ◽  
Times Series ◽  
Series Analysis ◽  
Data Source ◽  
The Times ◽  
Individual Value ◽  
Stock Values

A time series is a sequence of data values that are recorded with equal or varying time intervals. Time series data usually includes timestamps that indicate the time at which each individual value in the times series is recorded. Time series data is usually transmitted in the form of a data stream, i.e., continuous flow of data values. Source of time series data can be any system that measures and records data values over the course of time. Some examples of time series data may be recorded from stock values, blood pressure of a patient, temperature of a room, amount of a product in the inventory, and amount of precipitation in a region. Proper analysis and mining of time series data may yield valuable knowledge about the underlying characteristics of the data source. Time series analysis and mining has applications in many domains, such as financial, biomedical, and meteorological applications, because time series data may be generated by various sources in different domains.

Sign in / Sign up

Export Citation Format

Share Document