Search

Objective: The aim of this study was to predict the number of patients infected with coronavirus disease (COVID-19) by the end of year 2020 by using the number of infections in Turkey between the dates 11.3.2020-9.9.2020 and exponential smoothing model. Material and Methods: Different versions of the exponential smoothing model were applied and compared by using the data published and upgraded on the web by the Turkish Republic Ministry of Health, and near future predictions were performed after the determination of two models (containing additive method model: Model 1 and containing multiplicative method model: Model 2) producing the best estimates. Results: The first proposed model (Model 1) estimated that there would be 374,842, 427,672 and 482,263 infected patients in Turkey by the date 31 October 2020, 30 November 2020, and 31 December 2020, respectively. For the same dates, it was calculated that there would be 391,992, 472,782, and 573,811 infected patients, respectively for the second model (Model 2). The model 2 was evaluated to produce more realistic results considering that the number of patients infected with COVID-19 will increase even more with the effect of winter. Conclusion: It was thought that the models proposed in this study could also be used to predict the number of infected patients in international studies in the upcoming period. At the same time, it was evaluated that these models not only for infections but could also be used to estimate the number of deaths and recoveries due to coronavirus disease 2019. [ABSTRACT FROM AUTHOR]

In this research, smoothing methods and the Ordinary Least Squares (OLS) method for predicting producer milk prices in Turkey is implemented. Holt-Winters multiplicative (HWM) method and Holt-Winters additive (HWA) method are included in the exponential smoothing methods. Producer milk prices data belong to January 1st, 2010 to December 31st, 2019 period and gathered from the Turkish Statistical Institute. According to the given results, OLS method is fitted the data and has good root mean square error (RMSE) and determination coefficients (R²). [ABSTRACT FROM AUTHOR]

In this paper, the first study using exponential smoothing methods and the Box-Jenkins method for forecasting consumer potato prices in Turkey is conducted. The exponential smoothing methods include the Holt-Winters multiplicative (HWM) method and the Holt-Winters additive (HWA) method. This study explores the application of the Holt-Winters approach to time series data of consumer potato prices covering the period from January 1st, 2005 to July 31st, 2019. The data were obtained from the database of the Turkish Statistical Institute. The paper explores and forecasts the trend of potato prices from August to December 2019. The results illustrate that the ARIMA method achieves good consumer potato price forecasting accuracy according to the mean absolute percentage error (MAPE), the root mean square error (RMSE) and the mean absolute deviation (MAD). As a result, the forecasted values from August 2019 to December 2019 are calculated in this paper via these methods. [ABSTRACT FROM AUTHOR]

The purpose of the present study is to develop models for the time series data of honey prices in Turkey between 1998 and 2018 using Box-Jenkins and Winter's Exponential Smoothing methods and to evaluate the TRY- and US$-denominated forecasts made using such models for 2019-2020. The method employed to analyse the time series data is the Seasonal Auto Regressive Integrated Moving Average (SARIMA). The most suitable model to forecast the honey prices was found to be SARIMA (2,1,2)(1,0,1) for TRY-denominated prices and SARIMA (0,1,3)(0,0,0) for US$-denominated prices. It was forecast that the average price of 1 kg of honey would be 49.08 TRY/6.15 US$ at the end of 2020. Since the study will provide beekeepers with information on future prices, it will help them better plan their production activities. The price forecasts in this study will be useful for producers and those involved in sale of honey with respect to prediction of how the honey prices will change in the years to come. [ABSTRACT FROM AUTHOR]

Oral and dental diseases and the fight against these diseases are a difficult and costly issue for our country and most of the world countries. Protective and informative dentistry practices may be realized with the prospective realistic planning of Dentist personnel trained at adequate number and Oral Dental Health services. Attaching importance to the protective services for development of health services are the leading among the issues, which the policy-makers are required to emphasize with caution. The aim of this study is to model the series of number of individual per dentist personnel in Turkey between the years of 1951-2016 with Box-Jenkins and Brown Exponential Smoothing methods, and to provide these forecast models derived to guide the policymakers. Number of individual per dentist personnel is calculated prospectively in this study and values expected until 2023 are found with Box Jenkins and Brown Exponential Smoothing methods frequently used in making prospective forecasts. Calculations of these two models are compared in terms of decision-making criteria and it is concluded that Brown Exponential Smoothing model produces more coherent results than Box Jenkins model on behalf of making prospective forecasts. [ABSTRACT FROM AUTHOR]

Inversion of electromagnetic induction data is often realized using smoothing operators for regularizing the process. These operators are applied using derivative matrices, penalizing differences between neighboring mesh cells. The resulting smoothing effect is similar to that of low-pass Gaussian filters and may cause difficulties due to the smoothed boundaries between structures. To prevent such ambiguities, non-linear rank filters are applied widely in image processing applications to filter low frequency variations while keeping boundaries. Correspondingly, we have defined a non-linear rank order smoothing operator for regularization. The effects of the defined parameters are demonstrated on synthetic models and the operator is observed to be able to provide better defined boundaries. The implementation of the operator on field data is realized using single-frequency VLF-R data collected on profiles traversing the surface rupture of the North Anatolian Fault in Kocaeli, Turkey. In the study area, ERT data are also collected on a single profile, and its smooth inversions are realized individually and jointly with VLF-R data for comparisons. Inversions show that the recovered subsurface structure using the rank order smoothing is closer to that indicated by the ERT data and previous studies. According to the recovered models, the clayey fillings in the fault rupture are acting as a barrier affecting the horizontal fluid flow in the near-surface and the resulting water accumulations around the fault rupture may cause susceptibility for earthquake induced damages. • VLF-R data is inverted using a smoothing operator based on rank order filters. • The operator is observed to be able to provide models with better defined boundaries. • VLF-R data collected on the rupture of the North Anatolian Fault is inverted. • Inversions suggest, the clayey fills in the rupture causes water accumulations around the fault. [ABSTRACT FROM AUTHOR]

The objective of this study is to model the per capita consumption of red meat in Turkey employing various time series methods, evaluate the forecasting capability of the developed models, and address the variables that may affect the per capital consumption of red meat using the cointegration method on a term basis (short/long). The material of the study consists of the per capita consumption of red meat, total annual population, feed prices, gross domestic product and share of agriculture in gross domestic product in Turkey between 1993 and 2017. ARIMA (0,1,0) and Brown's exponential smoothing method were employed to model the time series data for per capita consumption of red meat, and Johansen method was used to address the cointegration relationship between per capita consumption of red meat and the other variables. The results of the modelling work suggest that per capita consumption of red meat has an increasing trend. Additionally, a statistically significant short-term relationship was found between per capita consumption of red meat and the other variables. Given the relationship between consumption of red meat and level of economic development, the projections concerning red meat consumption are important from the viewpoint of the policies that will be formulated. [ABSTRACT FROM AUTHOR]

The aim of the investigation was to forecast annual production of some oil seed crops (sesame, sunflower and soybean) in Turkey for the years 2016 through 2025 using annual production data for the period 1950-2015 and to give solid recommendation on production for producers, consumers and input providers. For this aim, three exponential smoothing methods, Holt, Brown and Damped Trend were executed to economically model the time series data. Goodness of fit criteria such as stationary R2, R2 and BIC criteria were adopted in the comparison of these exponential smoothing methods. Soybean, sunflower and sesame production amounts for the period 2016 -2025 were forecasted with high accuracy by using Holt exponential smoothing method with two parameters, which yielded the best result among exponential smoothing methods. Forecasted production amounts of soybean, sunflower and sesame from the period 2016- 2025 ranged from 162.878 to 179.784, 1.692.269 to 1.879.521 and 18.212 to 15.318 tons, respectively. We hope that the results from the time series data will provide baseline information for sustaining production and for guiding agricultural policy and exports of Turkey in terms of the above-mentioned plants in forthcoming years. [ABSTRACT FROM AUTHOR]