Seasonal Autoregressive Integrated Moving Average Model Research Articles

Ten daily rainfall forecasting using SSA algorithms and Seasonal ARIMA model to determine the beginning of the rainy season

Rainfall is one of the most important hydrological parameters. Based on the prediction of rainfall, several plans related to rainfall can be carried out, including predicting the start of the planting season based on information on the beginning of the rainy season and preparing cropping patterns. The purpose of this study is to predict the ten daily rainfall for 12 months from January to December of the following year and predict the start of the rainy season based on BMKG criteria. Modeling and forecasting is carried out on the ten daily regional rainfall in the catchment of Citarum-Majalaya water gauge using the Singular Spectrum Analysis (SSA) algorithm and the Seasonal ARIMA model. Forecasting model performance is measured using error values, namely mean absolute percentage error (MAPE). The results of these performance tests illustrate that the SSA algorithm and the Seasonal ARIMA model have fairly good accuracy for predicting rainfall for the next 6 months, where each has a MAPE value of 36.8 percent and 40.0 percent. Prediction of the beginning of the rainy season at the study location based on SSA forecasting results is most likely to occur on the 3rd of October, in accordance with BMKG predictions. whereas based on the results of the SARIMA forecasting model, the beginning of the rainy season is likely to occur on the 1st of November, and another possibility that the beginning of the rainy season will occur on the 2nd of October.

The prediction and analysis of global climate change based on SARIMA

Global climate change is a significant challenge that the world is currently facing. Accurate prediction of global climate change is essential for environmental protection, agricultural production, and social development. This study explores the utilization of the Seasonal Autoregressive Integrated Moving Average (SARIMA) model for forecasting global climate change. The SARIMA model is a machine learning algorithm that can effectively capture seasonal patterns and non-linear characteristics of climate data. The study initiates by performing data preprocessing tasks, which encompass data cleaning, managing missing values, and converting the data into a suitable format for analysis. The SARIMA model is then constructed, considering the seasonality and autocorrelation of the climate data. Historical climate data is used to train the SARIMA models, which are then utilized to forecast future global climate changes. The predictive performance of the models is evaluated to validate the effectiveness and accuracy of the SARIMA model in global climate change prediction. Experimental results indicate that the SARIMA model effectively captures the underlying patterns and dynamics of the climate data. The accurate predictions of the SARIMA model have practical implications for understanding and forecasting global climate change. These forecasts provide insightful information for policy formulation and decision-making, aiding in the development of innovative strategies to mitigate and adapt to climate change.

A SARIMA time series forecasting for dengue cases for reporting to Yangon Region, Myanmar

Dengue fever is a significant public health challenge in Myanmar, which requires accurate monitoring to mitigate its impact. The study aimed to develop a forecasting model for dengue cases in Myanmar's Yangon region using historical data from January 2002 to December 2022, with the objective of enhancing epidemiological surveillance and outbreak management. This retrospective observational study examines dengue cases in Yangon from January 2002 to December 2022, employing Seasonal Autoregressive Integrated Moving Average (SARIMA) models for predictive analysis. The most accurate model identified was SARIMA (2,0,1) (1,1,1) 12, with an AIC (Akaike Information Criterion) of 206.19 and MAPE (Mean Absolute Percentage Error) of 1.47%. According to the model, a peak in dengue cases was expected in July 2023, with an estimated 451 cases between January and December that year. Spatial variations in dengue incidence across Yangon's townships emphasize the need for targeted interventions. While the SARIMA model is valuable, it would also be important to consider many other risk factors like climate, migration patterns, virus characteristics, and socioecological factors to improve forecasting accuracy. These findings can aid public health policymakers in preventing and managing dengue outbreaks in Myanmar. However, additional research is needed to incorporate additional risk factors into the model to comprehensively understand dengue epidemiology and improve forecasting accuracy.

Modeling of the COVID-19 impact on air passenger traffic in the US, European countries, and China

The COVID-19 pandemic has changed many aspects of people's lives including travel since early 2020. Specifically, it has adversely affected people traveling by air and has hit the air transport industry significantly. But, how big is the COVID-19 impact? In order to answer such a question, we collected air passenger traffic data from the US, European countries, and China which accounted for over 75% of the world's total air passenger traffic. Air passenger traffic data in these three regions during the period January 2010 to December 2019 were modeled using seasonal autoregressive integrated moving average (ARIMA) models. Seasonal ARIMA models were used to predict air passenger traffic from January 2011 to December 2019 (just before the spread of COVID-19) and the accuracy of the models was evaluated. The models were then used to predict air passenger traffic from January 2020 to December 2022 for the case without COVID-19. The COVID-19 impacts on air passenger traffic were estimated by calculating the differences in predicted and actual air passenger numbers in monthly basis. Results showed that air passenger traffic was significantly recovered in the US and European countries but it encountered significant falls in 2021 and 2022 in China due to spikes in COVID-19 variant cases in many provinces and the implementation of zero-tolerance COVID-19 policy. Implications of the study are given.

Trend and Seasonality of Hip Fractures in Catalonia, Spain: Exploring the Influence of Climate

To describe the secular trend and seasonality of the incidence of hip fracture (HF) and its relationship with climatic variables during the period 2010–2019 in Catalonia in people aged ≥ 65 years. The results were analyzed by sex, age groups (65–74, 75–84, and ≥ 85), and types of fracture (extracapsular and intracapsular). Data on sex, age, type of fracture, year, and month of hospitalization of patients admitted with a diagnosis of HF between January 1, 2010 and December 31, 2019 were collected. Crude and standardized HF incidence (HFi) rates were obtained. Data on the monthly mean of climatological variables (temperature, insolation, icy days, rain, relative humidity, atmospheric pressure, and wind force) were obtained from the network of meteorological stations in Catalonia. Time series analytical statistics were used to identify trends and seasonality. Linear regression and a seasonal autoregressive integrated moving average (ARIMA) were used to analyze the relationship of each climatic parameter with fracture rates. In addition, generalized additive models were used to ascertain the best predictive model. The total number of HF episodes was 90,149 (74.1% in women and 25.9% in men). The total number of HFs increased by 6.4% between 2010 and 2019. The median age (SD) was 84.5 (7.14) and 54% of patients were ≥ 85 years of age. Extracapsular fractures were the most common (55%). The standardized incidence rates decreased from 728.1/100,000 (95% CI 738.6–769.3) to 624.5/100,000 (95% CI 648.7–677.0), which represents a decrease of 14.2% (p < 0.05). The decline was greater at older ages. There were seasonal variations, with higher incidences in autumn (27.2%) and winter (25.7%) and lower rates in summer (23.5%) and spring (23.6%). Seasonality was more pronounced in elderly people and men. In the bivariate regression analysis, high temperatures and greater insolation were negatively associated with the HF rate, while the number of icy days, rainy days, and high relative humidity were associated with a higher incidence of fractures in all age groups and sexes. In the regression analysis using the seasonal ARIMA model, only insolation had a consistently significant association with overall HFi, after adjusting by trend and other climatic parameters. While the global number of HFs grew in Catalonia due to increases in the elderly population, the standardized HF rate decreased during the years 2010–2019. There was a seasonal trend, with predominance in the cold months and correlations with climatic parameters, especially with insolation.

Forecasting seasonal influenza activity in Canada-Comparing seasonal Auto-Regressive integrated moving average and artificial neural network approaches for public health preparedness.

Public health preparedness is based on timely and accurate information. Time series forecasting using disease surveillance data is an important aspect of preparedness. This study compared two approaches of time series forecasting: seasonal auto-regressive integrated moving average (SARIMA) modelling and the artificial neural network (ANN) algorithm. The goal was to model weekly seasonal influenza activity in Canada using SARIMA and compares its predictive accuracy, based on root mean square prediction error (RMSE) and mean absolute prediction error (MAE), to that of an ANN. An initial SARIMA model was fit using automated model selection by minimizing the Akaike information criterion (AIC). Further inspection of the autocorrelation function and partial autocorrelation function led to 'manual' model improvements. ANNs were trained iteratively, using an automated process to minimize the RMSE and MAE. A total of 378, 462 cases of influenza was reported in Canada from the 2010-2011 influenza season to the end of the 2019-2020 influenza season, with an average yearly incidence risk of 20.02 per 100,000 population. Automated SARIMA modelling was the better method in terms of forecasting accuracy (per RMSE and MAE). However, the ANN correctly predicted the peak week of disease incidence while the other models did not. Both the ANN and SARIMA models have shown to be capable tools in forecasting seasonal influenza activity in Canada. It was shown that applying both in tandem is beneficial, SARIMA better forecasted overall incidence while ANN correctly predicted the peak week.

Deep Learning Model Effectiveness in Forecasting Limited-Size Aboveground Vegetation Biomass Time Series: Kenyan Grasslands Case Study

Timely forecasting of aboveground vegetation biomass is crucial for effective management and ensuring food security. However, research on predicting aboveground biomass remains scarce. Artificial intelligence (AI) methods could bridge this research gap and provide early warning to planners and stakeholders. This study evaluates the effectiveness of deep learning (DL) algorithms in predicting aboveground vegetation biomass with limited-size data. It employs an iterative forecasting procedure for four target horizons, comparing the performance of DL models—multi-layer perceptron (MLP), long short-term memory (LSTM), gated recurrent unit (GRU), convolutional neural network (CNN), and CNN-LSTM—against the traditional seasonal autoregressive integrated moving average (SARIMA) model, serving as a benchmark. Five limited-size vegetation biomass time series from Kenyan grasslands with values at 15-day intervals over a 20-year period were chosen for this purpose. Comparing the outcomes of these models revealed significant differences (p &lt; 0.05); however, none of the models proved superior among the five time series and the four horizons evaluated. The SARIMA, CNN, and CNN-LSTM models performed best, with the statistical model slightly outperforming the other two. Additionally, the accuracy of all five models varied significantly according to the prediction horizon (p &lt; 0.05). As expected, the accuracy of the models decreased as the prediction horizon increased, although this relationship was not strictly monotonic. Finally, this study indicated that, in limited-size aboveground vegetation biomass time series, there is no guarantee that deep learning methods will outperform traditional statistical methods.

Research on Automated Pricing and Supplementation Decision-making of Vegetable Products

In fresh supermarkets, the shelf life of vegetable products is usually very short, so it is crucial to replenish inventory daily based on historical sales data and product line requirements. In order to strengthen decision-making and inventory management, this article first selects representative supermarket sales data and conducts appropriate preprocessing, and then adopts a series of effective analysis methods. Firstly, using Kappa consistency testing to explore the correlation between different vegetable categories can help determine which products may be correlated in sales. In addition, in order to better formulate pricing and replenishment strategies for vegetable products, this article introduces a cost-plus pricing model. When determining the markup coefficient, multiple linear regression was used to ensure that the price-setting process is reasonable and consistent with cost and demand. In order to further confirm the accuracy and feasibility of this method, the article calculated the R-squared value of multiple linear regression, which reached 0.975, providing strong validation for the effectiveness of this method. When formulating replenishment strategies, in order to obtain sales forecasts for vegetable categories within the next 7 days, this article also uses a seasonal ARIMA model, which helps supermarkets plan inventory and supply chain more effectively. These strengthening measures have jointly improved the decision-making ability of vegetable product management in supermarkets, promoted more effective data-driven strategies, ensured the freshness and quality of vegetable products, and maximized the economic benefits of supermarkets.

Trend analysis and prediction of gonorrhea in mainland China based on a hybrid time series model

BackgroundGonorrhea has long been a serious public health problem in mainland China that requires attention, modeling to describe and predict its prevalence patterns can help the government to develop more scientific interventions.MethodsTime series (TS) data of the gonorrhea incidence in China from January 2004 to August 2022 were collected, with the incidence data from September 2021 to August 2022 as the validation. The seasonal autoregressive integrated moving average (SARIMA) model, long short-term memory network (LSTM) model, and hybrid SARIMA-LSTM model were used to simulate the data respectively, the model performance were evaluated by calculating the mean absolute percentage error (MAPE), root mean square error (RMSE), and mean absolute error (MAE) of the training and validation sets of the models.ResultsThe Seasonal components after data decomposition showed an approximate bimodal distribution with a period of 12 months. The three models identified were SARIMA(1,1,1) (2,1,2)12, LSTM with 150 hidden units, and SARIMA-LSTM with 150 hidden units, the SARIMA-LSTM model fitted best in the training and validation sets, for the smallest MAPE, RMSE, and MPE.ConclusionsThe overall incidence trend of gonorrhea in mainland China has been on the decline since 2004, with some periods exhibiting an upward trend. The incidence of gonorrhea displays a seasonal distribution, typically peaking in July and December each year. The SARIMA model, LSTM model, and SARIMA-LSTM model can all fit the monthly incidence time series data of gonorrhea in mainland China. However, in terms of predictive performance, the SARIMA-LSTM model outperforms the SARIMA and LSTM models, with the LSTM model surpassing the SARIMA model. This suggests that the SARIMA-LSTM model can serve as a preferred tool for time series analysis, providing evidence for the government to predict trends in gonorrhea incidence. The model's predictions indicate that the incidence of gonorrhea in mainland China will remain at a high level in 2024, necessitating that policymakers implement public health measures in advance to prevent the spread of the disease.

Seasonal Autoregressive Integrated Moving Average (SARIMA) for Melon (Cucumis melo) Yield from 2011 to 2020 Based on Planting Date Period

The main goal of time series modelling is to collect and analyze past values to develop appropriate models that describe the inherent structure and characteristics of the series about the planting date of melon. The data used in this study are the melon yield based on twenty-nine days of planting date in March from the first to the last planting date of March 2011 to 2020, to evaluate the prediction performances of the models, two indices: the root mean squared error (RMSE) and mean absolute percentage error (MAPE) were used to compare the forecasting capabilities of the SARIMA models, Excel was used first for preparing as CSV file and then changed to time series data to make the dataset of yield and R version 4.0.2 (the R Development Core Team) was used to perform ETS and SARIMA models For planting date time series based on yield, seasonal autoregressive integral moving average models (SARIMA) were constructed. The Student's t-test was one of the statistical tests used to evaluate the effectiveness and validity of the SARIMA models Through the use of SARIMA models, it is feasible to create synthetic records that maintain the statistical properties of the historical record. Finally, the results can be applied to different planting dates based on yield. Based on the Bayesian Information Criteria (BIC) values and the overall highest R2 values of 0.94, the optimal SARIMA (0,1,1) model was chosen. No limitations were found. It is the first time that this mathematical method was used to analyze time series data for 10 years of planting date of melon and it could be useful for agronomists and horticulturists to choose the best model.

Forecasting the incidence of dengue fever in Malaysia: A comparative analysis of seasonal ARIMA, dynamic harmonic regression, and neural network models

Currently, no vaccines or specific treatments are available to treat or prevent the increasing incidence of dengue worldwide. Therefore, an accurate prediction model is needed to support the anti-dengue control strategy. The primary objective of this study is to develop the most accurate model to predict future dengue cases in the Malaysian environment. This study uses secondary data collected from the weekly reports of the Ministry of Health Malaysia (MOH) website over six years, from 2017 to 2022. Three forecasting techniques, including seasonal autoregressive integrated moving average (SARIMA), dynamic harmonic regression (DHR), and neural network autoregressive model (NNAR), were first fitted to the estimation part of the data. First, several SARIMA models were run, and the best seasonal model identified was SARIMA (0, 1, 2) (1, 1, 1)52. The best DHR model was obtained with a Fourier term of 2, as this corresponds to the lowest Akaike Information Criteria (AIC) value. The NNAR (9, 1, 6)52 was considered the best choice among the NNAR models due to its superior performance in terms of the lowest error measures. The comparison among the three techniques revealed that the DHR model was the best due to its lowest MAPE and RMSE values. Thus, the DHR model was used to generate future forecasts of weekly dengue cases in Malaysia until 2023. The results showed that the model predicted more than a thousand dengue cases around weeks 27 to 32. The results showed an increase in dengue cases after the end of the monsoon season, which lasted about five months. This technique is proving to be valuable for health administrators in improving preparedness.

A Hybrid GARCH and Deep Learning Method for Volatility Prediction

Volatility prediction plays a vital role in financial data. The time series movements of stock prices are commonly characterized as highly nonlinear and volatile. This study is aimed at enhancing the accuracy of return volatility forecasts for stock prices by investigating the prediction of their price volatility through the integration of diverse models. Thus, the study integrated four powerful methods: seasonal autoregressive (AR) integrated moving average (MA), generalized AR conditional heteroskedasticity (ARCH) family models, convolutional neural network (CNN), and bidirectional long short‐term memory (LSTM) network. The hybrid model was developed using the residuals generated by the seasonal AR integrated MA model as input for the generalized ARCH model. Following this, the estimated volatility obtained was utilized as an input feature for both the hybrid CNNs and bidirectional LSTM models. The model’s forecasting performance was assessed using key evaluation metrics, including mean absolute error (MAE) and root mean squared error (RMSE). Compared to other hybrid models, our new proposed hybrid model demonstrates an average reduction in MAE and RMSE of 60.35% and 60.61%, respectively. The experimental results show that the model proposed in this study has good performance and accuracy in predicting the volatility of stock prices. These findings offer valuable insights for financial data analysis and risk management strategies.

Time Series Prediction of Temperature Using Seasonal ARIMA and LSTM Models

Time Series Prediction of Temperature Using Seasonal ARIMA and LSTM Models

Using the ARIMA/SARIMA Model for Afghanistan's Drought Forecasting Based on Standardized Precipitation Index

Forecasting drought plays a vital role in strategic planning and the management of underground water supply. In this study, we utilized autoregressive integrated moving average (ARIMA) and Seasonal ARIMA (SARIMA) models to predict drought events in Afghanistan, based on the standardized precipitation index (SPI). We used monthly average precipitation data from 1991 to 2015 for model training, while data from 2016 to 2020 were employed for model validation. The results of the statistical analysis, which encompassed evaluating Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Mean Absolute Percentage Error (MAPE), indicated that among the SPI 3, SPI 6, SPI 9, SPI 12, and SPI 24, the SARIMA models applied to the SPI 24 demonstrated the most accurate forecasting performance with RMSE (0.1492), MAE (0.1039), and MAPE (22.3732%) compared to SPI 3, SPI 6, SPI 9, and SPI 12. Subsequently, the ARIMA/SARIMA models were employed to forecast drought events for the upcoming year. It’s noteworthy that this constitutes the first-ever statistical analysis of the drought index in Afghanistan. Therefore, the outcomes of this study can be applied across diverse sectors, including water resource management and environmental precautions.

COMPARATIVE PERFORMANCE OF ARIMA, SARIMA AND GARCH MODELS IN MODELLING AND FORECASTING UNEMPLOYMENT AMONG ASEAN-5 COUNTRIES

Unemployment, especially after the COVID-19 pandemic, is a critical issue for any country as it has economic and social ramifications. Consequently, forecasting unemployment becomes an essential task as it can guide government policy. Time series data are frequently influenced by outliers (unexpected events), and some outliers may exist with extreme observation to reduce the forecasting effectiveness of robust estimators. This study compared the performance of Autoregressive Integrated Moving Average (ARIMA), Seasonal Autoregressive Integrated Moving Average (SARIMA) and Generalised Autoregressive Conditional Heteroscedasticity (GARCH) models in modelling and forecasting unemployment rates during the COVID-19 pandemic among the ASEAN-5 countries. These countries include Malaysia, Singapore, Thailand, the Philippines and Indonesia. The monthly unemployment data from January 2010 to December 2021 were applied for all cases, except Thailand, until December 2020. Each adequate model from both forecasting mechanisms underwent forecasting. Their performance was compared based on root mean squared error (RMSE), mean absolute error (MAE), Theil inequality coefficient and symmetric mean absolute percentage error (SMAPE). Static forecasting from the ARIMA and SARIMA models was found to perform better than the GARCH model in modelling and forecasting the unemployment rate among ASEAN-5 countries during the pandemic period.

Bootstrapping State-Space Models: Distribution-Free Estimation in View of Prediction and Forecasting

Linear models, seasonal autoregressive integrated moving average (SARIMA) models, and state-space models have been widely adopted to model and forecast economic data. While modeling using linear models and SARIMA models is well established in the literature, modeling using state-space models has been extended with the proposal of alternative estimation methods to the maximum likelihood. However, maximum likelihood estimation assumes, as a rule, that the errors are normal. This paper suggests implementing the bootstrap methodology, utilizing the model’s innovation representation, to derive distribution-free estimates—both point and interval—of the parameters in the time-varying state-space model. Additionally, it aims to estimate the standard errors of these parameters through the bootstrap methodology. The simulation study demonstrated that the distribution-free estimation, coupled with the bootstrap methodology, yields point forecasts with a lower mean-squared error, particularly for small time series or when dealing with smaller values of the autoregressive parameter in the state equation of state-space models. In this context, distribution-free estimation with the bootstrap methodology serves as an alternative to maximum likelihood estimation, eliminating the need for distributional assumptions. The application of this methodology to real data showed that it performed well when compared to the usual maximum likelihood estimation and even produced prediction intervals with a similar amplitude for the same level of confidence without any distributional assumptions about the errors.

Day-Ahead Electricity Load Forecasting with Multivariate Time Series

The importance of day-ahead load forecasting cannot be overstated. Electricity load forecasting is highly important because it allows electric distribution utilities to increase their transmission efficiency and their revenues, increase the reliability of power supply, and correct decisions for future developments. In this paper, day-ahead load forecasting was studied using multivariate time series analysis. Traditional forecasting method such as seasonal autoregressive integrated moving average (SARIMA) was used. Machine learning algorithms such as Feed Forward Neural Network (FFNN) and Long Short Term Memory (LSTM) were also utilized to determine their applicability to short-term load forecasting. For SARIMA, only historical hourly load was needed while FFNN and LSTM required the addition of temperature data, hour of the day and day of the week, special events and previous hour load. In the prediction, FFNN and LSTM performed better with mean absolute percentage error (MAPE) of 1.80 and 1.75%, respectively, compared with SARIMA with a MAPE of 4.48%. The study demonstrated that machine learning like FFNN and LSTM outperformed the traditional SARIMA models, highlighting their effectiveness in applications such as short-term load forecasting in time series prediction.

ON THE PERFORMANCE OF SARIMA AND SARIMAX MODEL IN FORECASTING MONTHLY AVERAGE RAINFALL IN KOGI STATE, NIGERIA

Forecasting monthly rainfall is very important in Kogi state for better approach to flood management and also plays a pivotal role in agriculture which remains a significant factor in Nigeria’s economy. Advanced time series univariate models such as Seasonal Autoregressive Integrated Moving Average (SARIMA) models are usually employed in modelling and forecasting rainfall in Nigeria due to their non-linear pattern and spatiotemporal variation. Few studies have attempted to investigate the influence of other climatic factors in modelling and prediction of rainfall pattern. This study examines the performance of a univariate seasonal ARIMA and seasonal ARIMA which uses monthly temperature and relative humidity as exogenous factors otherwise known as SARIMAX model in forecasting monthly average rainfall in Lokoja, the capital of Kogi state. The study uses monthly data on rainfall, temperature and relative humidity spanning from 2010 to 2022 obtained from Nigeria Meteorological Agency NiMet, Lokoja station. The series were appropriately differenced to attain stationarity. The plots of autocorrelation function (ACF) and partial autocorrelation function (PACF) were used to select some tentative models whose parameters would be estimated. The most suitable SARIMA model [SARIMA was chosen based on maximum Coefficient of Determination , and the minimum Akaike information criterion (AIC). However, SARIMAX model outperformed SARIMA model based on the criteria earlier highlighted. SARIMAX model was therefore recommended for modelling and forecasting monthly average rainfall in Kogi state.

FORECASTING THE VALUE OF INDONESIA'S OIL AND GAS IMPORTS USING SEASONAL AUTOREGRESSIVE INTEGRATED MOVING AVERAGE MODEL

The value of Indonesia's oil and gas imports is a combination of the value of crude oil (petroleum), oil and natural gas products. Throughout 2021, the value of Indonesia's oil and gas imports reach US$ 25.53 billion or the equivalent of 382.95 trillion rupiah (estimated at US$ 1 = Rp. 15,000.00). The high demand for petroleum in Indonesia is due to the fact that petroleum is the main source of energy for daily life needs, especially for industrial, transportation and household needs. The requirment for oil imports is expected to increase along with the growth in Indonesia's population. Therefore, a step is needed to prevent an increase in the value of oil and gas imports in the coming period. One method of analysis that can be used is forecasting using the time series method with the Seasonal Autoregressive Integrated Moving Average (SARIMA) model. The SARIMA model is a time series method with data that has a seasonal pattern and the forecasting results will get a pattern similar to the previous data. The data used is data on the monthly value of oil and gas imports from January 2005 to December 2022 with totaling 216 data. This research aims to find the best model and predict the value of Indonesia's oil and gas imports in the next 12 periods with data test in 4 periods (Januari to April 2023). The best model for the results of this research is (2, 1, 0)(0, 1, 1)43 with a MAPE value of 13.90%. Based on the accuracy of the MAPE value, this percentage has good quality forecasting results.

Forecasting of neonatal mortality trend at a special new-born care unit in Odisha, India: A time-series analysis

New born mortality is a public health problem in the state of Odisha. Newborn mortality is a dynamic process and variations in mortality are observed temporally and seasonally, and also across health facilities. Prior knowledge of mortality burden can enable health system’s readiness in terms of resources allocation and timely intervention, thereby improving the chances of survival of sick newborns admitted in the hospitals. Hence, this study aimed to examine temporal trends of newborn mortality in a Special Newborn Care Unit of Saheed Laxman Nayak Medical College and Hospital (SLNMCH) in Odisha and forecast a short-term monthly projection.The Box-Jenkins approach was used to fit a seasonal autoregressive integrated moving average (SARIMA) model to the monthly recorded mortality among the hospitalized new borns in the SNCU during 2016-2020. The best-fit model for forecasting was found based on the Akaike Information Criterion.The time-series analysis revealed a modest upward trend in newborn mortality rate among SNCU admitted newborns, with peaks in the late winter and late summer months. The seasonal ARIMA (0,1,1)(1,1,1)12 model offered the best fit for time-series data. This model predicted the monthly percentage of mortality in SNCU admitted newborns in the range of 9% to 35% with respective 95% confidence interval for two years period (2021-2022).SARIMA models are useful for monitoring newborn mortality and provide an estimate of temporal trends and seasonality. The models are helpful for predicting occurrence of mortality in the SNCU of SLNMCH and could be useful for developing early warning systems. It may help in early detection, timely treatment, and prevention of serious complications in admitted sick newborns.