Mean Absolute Error Criterion Research Articles

The Multicollinearity Effect on the Performance of Machine Learning Algorithms: Case Examples in Healthcare Modelling

Background: The data extracted from various fields inherently consists of extremely correlated measurements in parallel with the exponential increase in the size of the data that need to be interpreted owing to the technological advances. This problem, called the multicollinearity, influences the performance of both statistical and machine learning algorithms. Statistical models proposed as a potential remedy to this problem have not been sufficiently evaluated in the literature. Therefore, a comprehensive comparison of statistical and machine learning models is required for addressing the multicollinearity problem. Methods: Statistical models (including Ridge, Liu, Lasso and Elastic Net regression) and the eight most important machine learning algorithms (including Cart, Knn, Mlp, MARS, Cubist, Svm, Bagging and XGBoost) are comprehensively compared by using two different healthcare datasets (including Body Fat and Cancer) having multicollinearity problem. The performance of the models is assessed through cross validation methods via root mean square error, mean absolute error and r-squared criteria. Results: The results of the study revealed that statistical models outperformed machine learning models in terms of root mean square error, mean absolute error and r-squared criteria in both training and testing performance. Particularly the Liu regression often achieved better relative performance (up to 7.60% to 46.08% for Body Fat data set and up to 1.55% to 21.53% for Cancer data set on training performance and up to 1.56% to 38.08% for Body Fat data set and up to 3.50% to 23.29% for Cancer data set on testing performance) among regression methods as well as compared to machine algorithms. Conclusions: Liu regression is mostly disregarded in the machine learning literature, but since it outperforms the most powerful and widely used machine learning algorithms, it appears to be a promising tool in almost all fields, especially for regression-based studies including data with multicollinearity problem.

Empirical analysis of dust and noise pollutants produced in the wood-CNC machining process.

In the manufacturing processes, consideration of sustainability is of particular importance. The current study is concerned with the influences of changing the process variables on the reduction of pollutions in the wood-CNC machining operation. Noise and dust are the studied pollutants in the present research work. Process variables include feed rate, spindle speed, step-over, and depth of cut, and the aim is to predict the behavior of aforementioned pollutants variations in the current process. The amounts of these harmful factors are measured based on existing standards. In order to analyze the findings, adaptive neuro-fuzzy inference system (ANFIS) and regression analysis methods have been employed, separately. The effects of process parameters on response variables have been comprehensively studied. The research findings demonstrated that for the present problem, ANFIS outcomes are more accurate. According to the mean absolute error (MAE) criterion, the prediction errors of ANFIS for noise and dust factors were computed to, in turn, 0.50 and 14.89. Meanwhile, the error values for prediction of noise and dust responses using regression analysis were calculated as 1.54 and 34.62, respectively.

Modeling and prediction of flame retardant performance of alkyl phosphinate using artificial neural network and adaptive neural fuzzy inference system

AbstractThe flame retardancy of polymer materials is closely linked to the types and compositions of flame retardants used. However, due to the complex relationships involved, it is difficult to quantify and predict flame retardancy using linear equations and models. Alkyl phosphinate flame retardants have emerged as promising candidates for their environmentally friendly, low toxicity, and excellent flame retardant properties. In this study, we aimed to develop a novel, low‐cost alkyl phosphinate flame retardant by predicting the quantitative impact of flame retardant composition on flame retardancy. To establish prediction models for the limiting oxygen index (LOI) value of alkyl phosphinate flame retardants, we utilized artificial neural network (ANN) and adaptive neural fuzzy inference system (ANFIS) programs, namely ANN, ANFIS‐grid partition (GP), and ANFIS‐subtractive clustering (SC). The prediction models were trained and validated, and their reliability was assessed based on the root mean square error (RMSE), coefficient of determination (R2), and mean absolute percentage error (MAPE) criteria. Our results indicate that the ANFIS‐SC prediction model exhibits the highest reliability, with high accuracy and ideal prediction results.

A Multi-Stream Deep Neural Network to Predict the Energy Consumption of Smart Home Appliances

A novel ensemble machine learning approach for predicting energy consumption in smart appliances is presented in this paper. The main objective is minimizing the number of necessary sensors and improving the prediction accuracy at the same time. The proposed method combines the Fuzzy C-Means method with a multi-stream deep neural network to achieve this goal. The method focuses on prediction accuracy and aims to extract the minimum number of essential features from the dataset corresponding to the required sensors. These selected features are then scatter-reduced and homogenized into subsets. Each subset is used to train a cluster-specific deep neural network designed exclusively for that subset. The final prediction is obtained by computing the fuzzy-weighted sum of these cluster-specific network outputs. Numerical results show that the proposed prediction method outperforms conventional methods in terms of root mean square error and mean absolute percentage error criteria, despite using fewer sensors. This improvement can be attributed to the reduced dataset scatter, which improves the learning speed and model performance. Furthermore, the fuzzy combination of the outputs improves the final prediction accuracy. Overall, the proposed approach provides a more cost-effective and accurate solution for predicting the energy consumed by smart appliances.

Using meta-learning to recommend an appropriate time-series forecasting model

BackgroundThere are various forecasting algorithms available for univariate time series, ranging from simple to sophisticated and computational. In practice, selecting the most appropriate algorithm can be difficult, because there are too many algorithms. Although expert knowledge is required to make an informed decision, sometimes it is not feasible due to the lack of such resources as time, money, and manpower.MethodsIn this study, we used coronavirus disease 2019 (COVID-19) data, including the absolute numbers of confirmed, death and recovered cases per day in 187 countries from February 20, 2020, to May 25, 2021. Two popular forecasting models, including Auto-Regressive Integrated Moving Average (ARIMA) and exponential smoothing state-space model with Trigonometric seasonality, Box-Cox transformation, ARMA errors, Trend, and Seasonal components (TBATS) were used to forecast the data. Moreover, the data were evaluated by the root mean squared error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and symmetric mean absolute percentage error (SMAPE) criteria to label time series. The various characteristics of each time series based on the univariate time series structure were extracted as meta-features. After that, three machine-learning classification algorithms, including support vector machine (SVM), decision tree (DT), random forest (RF), and artificial neural network (ANN) were used as meta-learners to recommend an appropriate forecasting model.ResultsThe finding of the study showed that the DT model had a better performance in the classification of time series. The accuracy of DT in the training and testing phases was 87.50% and 82.50%, respectively. The sensitivity of the DT algorithm in the training phase was 86.58% and its specificity was 88.46%. Moreover, the sensitivity and specificity of the DT algorithm in the testing phase were 73.33% and 88%, respectively.ConclusionIn general, the meta-learning approach was able to predict the appropriate forecasting model (ARIMA and TBATS) based on some time series features. Considering some characteristics of the desired COVID-19 time series, the ARIMA or TBATS forecasting model might be recommended to forecast the death, confirmed, and recovered trend cases of COVID-19 by the DT model.

Application of ARIMA (Autoregressive Integrated Moving Average) model to predict Rupiah selling exchange rate against US Dollar

Currency is a tool in the form of money that is accepted and valid and legal as a payment and economic transactions in a country. US dollar is benchmark for world currencies, so predicting rupiah against US dollar is important. The purpose of this study is to analyze the characteristics of daily selling rate Rupiah against US Dollar, determine best model, and make predictions selling rate of Rupiah against US Dollar. Data used is daily data on selling rate of Rupiah against US Dollar 20 November 2020 - 19 January 2023 with details data training 20 November 2020 - 20 November 2022 and data testing 21 November 2022 - 19 January 2023. The model used is Autoregressive Integrated Moving Averages (ARIMA). The best model was chosen based on the smallest Mean Squared Error (MSE), Mean Absolute Percentage Error (MAPE) and Akaike Information Criterion (AIC). From the analysis results, it is found that best model is ARIMA (2,1,2) because it has significant parameters, white-noise residuals and has smallest MSE and MAPE values. With ARIMA model (2,1,2) the forecasting results for January 20 2023 – January 31 2023 is obtained with highest selling price on January 30 2023 Rp.15,932.4 and smallest on January 20 2023 Rp.15,901.9 and the average Rp.15,919.4. Based on these results, exporters and importers can consider their business activities.

Performance Evaluation of Artificial Neural Network Modelling to a Ploughing Unit in Various Soil Conditions

Abstract The specific objective of this study is to find a suitable artificial neural network model for estimating the operation indicators (disturbed soil volume, effective field capacity, draft force, and energy requirement) of ploughing units (tractor disc) in various soil conditions. The experiment involved two different factors, i.e., (Ι) soil texture index and (ΙΙ) field work index, and included soil moisture content, tractor engine power, soil bulk density, tillage speed, tillage depth, and tillage width, which were linked to one dimensionless index. We assessed the effectiveness of artificial neural network and multiple linear regression models between the values predicted and the actual values using the mean absolute error criterion to test data points. When the artificial neural network model was applied, the mean absolute error values for disturbed soil volume, effective field capacity, draft force, and energy requirement were 69.41 m3·hr−1, 0.04 ha·hr−1, 1.24 kN, and 1.95 kw·hr·ha−1, respectively. In order to evaluate the behaviour of new models, the coefficient R 2 was used as a criterion, where R 2 values in artificial neural network were 0.9872, 0.9553, 0.9948, and 0.9718, respectively, for the aforementioned testing dataset. Simultaneously, R 2 values in multiple linear regression were 0.7623, 0.696, 0.492, and 0.5572, respectively, for the same testing dataset. Based on these comparisons, it was clear that predictions using the artificial neural network models proposed are very satisfactory.

A novel neural-evolutionary framework for predicting weight on the bit in drilling operations

This study compares the performance of artificial neural networks (ANN) trained by grey wolf optimization (GWO), biogeography-based optimization (BBO), and Levenberg–Marquardt (LM) to estimate the weight on bit (WOB). To this end, a dataset consisting of drilling depth, drill string rotational speed, rate of penetration, and volumetric flow rate as input variables and the WOB as a response is used to develop and validate the intelligent tools. The relevance test is applied to sort the strength of WOB dependency on the considered features. It was observed that the WOB has the highest linear correlation with the drilling depth and drill string rotational speed. After dividing the databank into the training and testing (4:1) parts, the proposed LM-ANN, GWO-ANN, and BBO-ANN ensembles are constructed. A sensitivity analysis is then carried out to find the most powerful structure of the models. Each model performs to reveal the relationship between the WOB and the mentioned independent factors. The performance of the models is finally evaluated by mean square error (MSE) and mean absolute error criteria. The results showed that both GWO and BBO algorithms effectively help the ANN to achieve a more accurate prediction of the WOB. Accordingly, the training MSEs decreased by 14.62% and 24.90%, respectively, by applying the GWO and BBO evolutionary algorithms. Meanwhile, these values were obtained as around 9.86% and 9.41% for the prediction error of the ANN in the testing phase. It was also deduced that the BBO performs more efficiently than the other technique. The effect of input variables dimension on the accuracy and training time of the BBO-ANN clarified that the most accurate WOB predictions are achieved when the model constructs with all four input variables instead of utilizing either three or two of them with the highest linear correlation. It was also observed that the training stage of the BBO-ANN model with four input variables needs a little more computational time than its training with either two or three variables. Finally, the accuracy of the BBO-ANN model for the WOB prediction has been compared with the multiple linear regression, support vector regression, adaptive neuro-fuzzy inference systems, and group method of data handling. The statistical accuracy analysis confirmed that the BBO-ANN is more accurate than the other checked techniques.

Modelling Health State Utilities as a Transformation of Time to Death in Patients with Non-Small Cell Lung Cancer.

When utilities are analyzed by time to death (TTD), this has historically been implemented by 'grouping' observations as discrete time periods to create health state utilities. We extended the approach to use continuous functions, avoiding assumptions around groupings. The resulting models were used to test the concept with data from different regions and different country tariffs. Five-year follow-up in advanced non-small cell lung cancer (NSCLC) was used to fit six continuous TTD models using generalized estimating equations, which were compared with progression-based utilities and previously published TTD groupings. Sensitivity analyses were performed using only patients with a confirmed death, the last year of life only, and artificially censoring data at 24 months. The statistically best-fitting model was then applied to data subsets by region and different EQ-5D-3L country tariffs. Continuous (natural) [Formula: see text] and [Formula: see text] models outperformed other continuous models, grouped TTD, and progression-based models in statistical fit (mean absolute error and Quasi Information Criterion). This held through sensitivity and scenario analyses. The pattern of reduced utility as a patient approaches death was consistent across regions and EQ-5D tariffs using the preferred [Formula: see text] model. The use of continuous models provides a statistically better fit than TTD groupings, without the need for strong assumptions about the health states experienced by patients. Where a TTD approach is merited for use in modelling, continuous functions should be considered, with the scope for further improvements in statistical fit by both widening the number of candidate models tested and the therapeutic areas investigated.

Comparative analysis of deep learning and classical time series methods to forecast natural gas demand during COVID-19 pandemic

ABSTRACT The lockdown measures implemented to contain the COVID-19 pandemic have had a considerable effect on the consumption of natural gas, which is closely linked to the economic growth of countries. Accurately forecasting natural gas demand is critical for making informed decisions in unprecedented and unexpected situations. This study aims to compare artificial learning-based algorithms and classical statistical time series models in predicting natural gas demand during the pandemic, using Turkey as a case study. Common time series prediction methods, including Autoregressive Integrated Moving Average (ARIMA), Nonlinear Autoregression Neural Network (NARNN), Support Vector Regression (SVR), and Long Short-Term Memory (LSTM), were utilized for this purpose. The impact of the pandemic on natural gas demand was analyzed by including 2-year natural gas consumption data since its onset. Root mean square error (RMSE), correlation coefficient (R), and mean absolute error (MAE) criteria were used as performance evaluation metrics to select the best model. The results confirmed that the deep-learning-based LSTM model provided better prediction accuracy than time-series benchmark models, with the lowest RMSE (9.442) and the highest R (0.997) values in the test dataset. Furthermore, the results were validated by statistical analysis using the Diebold-Mariano and Nemenyi tests.

Aplikasi Model ARIMA dalam Peramalan Data Harga Emas Dunia Tahun 2010-2022

Gold investment is one of the favorite investments during the Covid-19 pandemic because the price of gold is relatively volatile but shows an increasing trend. Savvy investors investing in gold need to be able to predict future opportunities. Therefore, price estimation is needed to develop a buying and selling strategy to maximize profits. The Autoregressive Integrated Moving Average (ARIMA) model is a suitable method for predicting time series data, so the best ARIMA model will be applied for forecasting world gold prices. The best ARIMA model is selected based on the Akaike Information Criterion (AIC) and Mean Absolute Percentage Error (MAPE) criteria. Monthly world gold price data for 146 periods are applied in this study and will be used to predict gold prices for the following six periods. ARIMA(0,1,1) is the best model obtained from the analysis results, with AIC and MAPE values ​​of 1264.731 and 11.972%, respectively. Forecasting results show that world gold prices will increase for the next periods.

Modeling canopy cover and aboveground net primary production using soil factors in rangelands of Ardabil province, Iran

<p>Canopy cover (CC) and the aboveground net primary production (ANPP) vary under the influence of various environmental factors. They underscore ecological sustainability under different environment-human interactions. Towards this, the present study aimed to model the CC and ANPP of different plant functional types (PFTs) and their total using the soil attributes in the northwest (Ardabil province) rangelands of Iran. According to ecoregions and plant types and environmental factors, sampling was taken at the peak stage of plant growth from 2016 to 2020 using 1-m2 plots. For each transect, a soil sample was taken and transferred to the soil laboratory and the various attributes were measured. Maps of soil attributes were prepared using the inverse distance weighting (IDW) method. Differences in CC and ANPP of PFTs among soil attributes were analyzed using the paired sample t-test. Linear multiple regression was used for modeling the soil attributes. Total CC and ANPP were prepared in two ways (i.e., regression and PFTs maps summing). The accuracy assessment of the maps was calculated by using the criteria of mean absolute error (MAE), mean deviation error (MDE), and root mean squared error (RMSE). The obtained results were acceptable (value < 7). The difference between the modeled and measured mean values of CC was equal to –0.03% for grasses, –0.01% for forbs, +0.03% for shrubs, 0% for total<sub>regression model</sub>, and +0.06 for total<sub>PFTs maps summing</sub>. Additionally, this difference in terms of the ANPP was equal to 0 kg/ha for grasses, +0.02 kg/ha for forbs, –0.09 kg/ha for shrubs, –0.02 for total<sub>regression model</sub>, and +0.02 kg/ha for total<sub>PFTs maps summing</sub>. The present results are applicable to managing the balance between supply and demand of rangeland products and they can be used from carbon management perspectives.</p>

Development of a linear–nonlinear hybrid special model to predict monthly runoff in a catchment area and evaluate its performance with novel machine learning methods

Accurate forecasting of runoff as an important hydrological variable is a key task for water resources planning and management. Given the importance of this variable, in the current study, a multivariate linear stochastic model (MLSM) is combined with a multilayer nonlinear machine learning model (MNMLM) to generate a hybrid model for the spatial and temporal simulation of runoff in the Quebec basin, Canada. Monthly hydrological data from 2001 to 2013, including precipitation and runoff data from nine stations and Normalized Difference Vegetation Index (NDVI) extraction of MODIS data, are applied as input to the proposed hybrid model. At the first step of the hybrid modeling, data normality and stationary were examined by performing various tests. In the second step, MLSM was developed by defining four different scenarios and as a result 15 sub-scenarios. The first and second scenarios were developed based on one exogenous variable (precipitation or NDVI). In contrast, the second and third scenarios were developed based on two additional variables. In the first and third scenarios, the data are modeled without preprocessing. In the second and fourth scenarios, a preprocessing step is performed on the data. Then, in the third step, various combinations based on different time delays from runoff data were applied for developing nonlinear model. The comparisons are made between observed and simulated time series at various stations and based on the root mean squared error (RMSE), mean absolute error (MAE), correlation coefficient (R) and Akaike information criterion (AIC). The efficiency of the proposed hybrid model is compared with a novel machine learning model that was introduced in 2021 by Sultani et al., and it was also compared with the results obtained from the linear and nonlinear models. In most stations, delays (t-1) and (t-24) are identified as the most effective delays in hybrid and nonlinear modeling of runoff. Also, in most stations, the use of climatic parameters and physiographic factors as exogenous variables along with runoff data improves the results compared to the use of one variable. Results showed that at all stations, proposed hybrid model generally leads to more accurate estimates of runoff compared with various linear and nonlinear models. More accurate estimates of peak runoff values at all stations were another excellence of proposed hybrid model than other models.

A novel missing value imputation relying on K-means clustering and kernel-based weighting using grey relation (KWGI)

Data pre-processing is one of the crucial phases of data mining that enhances the efficiency of data mining techniques. One of the most important operations performed on data pre-processing is missing values imputation in incomplete datasets. This research presents a new imputation technique using K-means and samples weighting mechanism based on Grey relation (KWGI). The Grey-based K-means algorithm applicable to all samples of incomplete datasets clusters the similar samples, then an appropriate kernel function generates appropriate weights based on the Grey relation. The missing values estimation of the incomplete samples is done based on the weighted mean to reduce the impact of outlier and vague samples. In both clustering and imputation steps, a penalty mechanism has been considered to reduce the similarity of ambiguous samples with a high number of missing values, and consequently, increase the accuracy of clustering and imputation. The KWGI method has been applied on nine natural datasets with eight state-of-the-art and commonly used methods, namely CMIWD, KNNI, HotDeck, MeanI, KmeanI, RKmeanI, ICKmeanI, and FKMI. The imputation results are evaluated by the Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) criteria. In this study, the missing values are generated at two levels, namely sample and value, and the results are discussed in a wide range of missingness from low rate to high rate. Experimental results of the t-test show that the proposed method performs significantly better than all the other compared methods.

2D and 3D numerical simulations of dam-break flow problem with RANS, DES, and LES

In this study, dam-break flows occurring on different downstream/upstream water depth ratios (α = 0, 0.1, and 0.4) and different obstacle shapes are numerically modeled in 2D and 3D. The Finite Volume Method (FVM) is used in the numerical solution of the basic equations governing the propagation wave formed as a result of the dam-break, and the interface of water with air is calculated by the Volume of Fluid (VOF) method. In modeling turbulence stresses, four turbulence models particularly successful in curvilinear trajectories and high strain flows, namely the Re-normalization Group (RNG), Shear Stress Model (SST), Detached Eddy Simulation (DES), are employed for the 2D simulation, and Large Eddy Simulation (LES) for the 3D simulation. The numerical results are compared against the previous experimental results and the performance of the turbulence models is evaluated according to the Mean Absolute Relative Error (MARE) criteria. For the 2D analysis, it has been determined that the DES is more successful for α = 0, 0.1, and 0.4, whereas the SST is more successful for triangular and trapezoidal obstacles in the downstream region than other turbulence models. Comparing the 2D and 3D results, the 3D LES model is expected to be more compatible with the experimental water surface profiles for all conditions among the models used in the study. It has been concluded that turbulence models, in which the strain tensor is considered, are successful in modeling the propagation wave formed as a result of a dam-break.

Statistical modelling and forecasting annual sugarcane production in India: Using various time series models

AbstractThis paper proposes a comparison of various time series forecasting models to forecast annual data on sugarcane production over 63 years from 1960 to 2022. In this research, the Mean Forecast Model, the Naive Model, the Simple Exponential Smoothing Model, Holt's model, and the Autoregressive Integrated Moving Average time series models have all been used to make effective and accurate predictions for sugarcane. Different scale‐dependent error forecasting techniques and residual analysis have been used to examine the forecasting accuracy of these time series models. SE of Residuals, Root Mean Squared Error (RMSE), Mean Absolute Error (MAE) and Akaike's Information Criterion (AIC) are used to analyse the forecast's accuracy. The best model has been selected based on the predictions with the lowest value, according to the three‐performance metrics of RMSE, MAE, and AIC. The estimated sugarcane production shows an increasing trend for the next 10 years and is projected to be 37,763.38 million tonnes in the year 2032. Further, empirical results support the plan and execution of viable strategies to advance sugarcane production in India to fulfil the utilisation need of the increasing population and further improve food security.

RESEARCH ON THE PRINCIPLES OF ARTIFICIAL NEURAL NETWORK CONSTRUCTION AND PARAMETER MODELING

Studies show that neural networks with multilayer direct propagation have be-come more widespread, which is due to the simplicity of their implementation, the availability of advanced learning methods, and the parallel execution of calculations. In these networks, the computational complexity increases proportionally to the square of the data size, which leads to a decrease in performance and an increase in hardware costs. The main problems of synthesis of artificial neural networks, criteria for evaluating the effectiveness of artificial neural networks, issues of choosing layers and the number of neurons in each layer are investigated. In the work, the mean square error, mean absolute error and mean absolute percentage error criteria are considered for evaluating the efficiency of neural networks. In general, one hidden layer of a neu-ral network is sufficient for most problems, two hidden layers can represent a function of any shape, and adding a third and more layers results in very little improvement in network performance. Expressions are given for the total number of parameters de-pending on the number, and the number of inputs from the binary sequence, as well as the minimum and maximum values of the training examples depending on the total number of parameters and the number of outputs. The studies have shown that the se-lection of the statistical model efficiency evaluation criteria depends on the problem to be solved. The results of the simulations performed using the MATLAB package are presented.

Analysis and Forecasting of Rice Production Behavior in Bangladesh

This research looks at how numerous factors, including the price of rice, area, improved seed distribution, areas irrigated by different improved methods, and fertilizer, influence rice production in Bangladesh. This study employed the autoregressive distributed lag (ARDL) model to investigate the short-run and long-run effects of each exogenous variable on rice production. In this research work, auto regressive integrated moving average (ARIMA) and hybrid ARIMA-(general autoregressive conditional heteroskedastic) GRACH were used to forecast rice production in Bangladesh from 1983 to 2030. The parameters of price of rice, area irrigated by different improved methods, area, and fertilizer were found to have a significant and positive impact on rice production in this study, both in the short run and the long run, according to the ARDL model. According to the lowest Root Mean Squared Error, Mean Absolute Error and Theil Inequality Coefficient criteria, the linear hybrid ARIMA (11, 1, 11)–EGARCH (1, 1)–M model performed better in predicting the rice production in Bangladesh. IUBAT Review—A Multidisciplinary Academic Journal, 5 (1): 1-14

Forecasting blood demand for different blood groups in Shiraz using auto regressive integrated moving average (ARIMA) and artificial neural network (ANN) and a hybrid approaches

Providing fresh blood to keep people in need of blood alive, has always been a main issues of health systems. Right policy-making in this area requires accurate forecasting of blood demand. The current study aimed at predicting demand for different blood groups in Shiraz using Auto Regressive Integrated Moving Average (ARIMA), Artificial Neural Network (ANN) and a hybrid approaches. In the current time series analysis, monthly data of the Shiraz hospitals and medical centers demand for 8 blood groups during 2012–2019 were gathered from Shiraz branch of Iranian Blood Transfusion Organization. ARIMA, ANN and a hybrid model of them was used for prediction. To validate and comprise ARIMA and ANN models, Mean Square Error (MSE) and Mean Absolute Error (MAE) criteria were used. Finally, ARIMA, ANN and hybrid model estimates were compared to actual data for the last 12 months. R3.6.3 were used for statistical analysis. Based on the MSE and MAE of models, ARIMA had the best prediction for demand of all blood groups except O+ and O−. Moreover, for most blood groups, ARIMA had closer prediction to actual data. The demand for four blood groups (mostly negative groups) was increasing and the demand for other four blood groups (mostly positive ones) was decreasing. All three approaches including ARIMA, ANN and the hybrid of them predicted an almost downward trend for the total blood demand. Differences in the performance of various models could be due to the reasons such as different forecast horizons, daily/month/annual data, different sample sizes, types of demand variables and the transformation applied on them, and finally different blood demand behaviors in communities. Advances in surgical techniques, fetal screening, reduction of accidents leading to heavy bleeding, and the modified pattern of blood request for surgeries appeared to have been effective in reducing the demand trend in the current study. However, a longer time period would certainly provide more accurate estimates.

TIME SERIES MODELING OF NATURAL GAS FUTURE PRICE WITH FUZZY TIME SERIES CHEN, LEE AND TSAUR

Investment is the process of investing money or capital for profit or material results. The investor carefully calculates the investment object to minimize losses and maximize profits. One of the essential investment objects is the futures price of natural gas considered a commodity that plays a vital role in the Indonesian economy. The movement of natural gas futures prices can be modeled using a time series model. The data in the time series model is believed to have particular pattern to model the data in the future. The natural gas futures price is modeled into a time series method by using fuzzy time series (FTS) approach of the FTS Chen, Lee and Tsaur. Model accuracy is calculated using the criteria of Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Mean Absolute Percentage Error (MAPE). The three FTS methods have good performance of accuracy for this time series data, where FTS Tsaur as fuzzy times series approach with average based method shows the best results with the smallest error rate to the data of natural gas future price.