Content uploaded by Gaurav Mishra
Author content
All content in this area was uploaded by Gaurav Mishra on Nov 19, 2022
Content may be subject to copyright.
A preview of the PDF is not available
Modeling of rainfall time series using NAR and ARIMA model over western Himalaya, India
Abstract
The high Himalayas in northern India are an essential source of climate generation and maintenance over the entire northern belt of the Indian subcontinent. It also affects extreme weather phenomena such as western disturbances in the region during winter. The work presented here describes the trends in 117-year precipitation changes and their impact on the western Himalayas and suggests some possible explanations in the context of changing rainfall patterns. Under the investigation, the forecasting efficiency and the prediction pattern of artificial neural network (ANN) and seasonal autoregressive integrated moving average (SARIMA) models for rainfall series in the western Himalayan states of India have been assessed. The results revealed significant changes in the monthly, seasonal, and annual rainfall series data for the three states of the Western Himalayan regions from the years 1900 to 2017. The study also concludes that the nonlinear autoregressive neural network (NARNN) models can be used to forecast the western Himalayan region data series well. According to the result interpretation, the highest rainfall may be estimated in August, 1632.63 mm (2023), whereas the lowest rainfall can be obtained in October (0.43 mm) during 2023. The model predicted a gradual decrease in annual rainfall trends in Uttarakhand and Himachal Pradesh from 2018 to 2023 despite heavy rainfall prediction in the monsoon season, whereas Jammu and Kashmir increase in annual rainfall has been predicted from 2018 to 2023. Possible explanations for the change in precipitation over the western Himalayas have also been proposed and explained.
Find the full paper - https://rdcu.be/cZ2fl
... As supermarkets continue their pursuit of operational excellence, this study holds the potential to yield valuable insights that can be applied in the development and implementation of automated pricing and replenishment decision-making systems within the realm of vegetable goods. It is worth noting that the extensive data utilized in this paper has been sourced from the distinguished 2023 National College Students Mathematical Modeling Competition Question C [6]. (1) ...
... In this paper, we take a daily sales volume of large data -amaranth as a model cited example, and finally, we automatically find the optimal parameters based on the AIC information criterion, the model results for the ARIMA model (1, 0, 3) test table, based on the variables: commodity 102900005115762_min-max standardization, from the Q The analysis of the results of the statistics can be obtained: Q6 does not show significance in the level, cannot reject the hypothesis that the residuals of the model are white noise series, at the same time, the model's goodness of fit R² is 0.679, the model performs better, the model meets the requirements. The model formula is as follows: (6) After the inverse normalization process, the fitted image is derived with the prediction as shown in Figure 5 below: ...
In a fresh food superstore, the superstore replenishes daily based on the historical sales and demand of vegetable commodities. To formulate the optimal replenishment decision, this paper analyses the sales flow, purchase cost, and loss of a supermarket in a period, and through the use of a variety of models, it can give a reference opinion on the formulation of automatic pricing and replenishment decisions for vegetable commodities. The first text describes the process of analyzing the relationship between total sales volume and cost-plus pricing, including data integration, preprocessing, and multiple regression analysis methods. Through these analyses, a general relationship between total sales volume and merchandise selling price is identified. Univariate linear regression, polynomial regression, random forest regression, and BP neural network regression models were all used to analyze this relationship. The focus is then on the total daily replenishment and pricing strategy for the coming week to maximize the superstore's revenue. The text details the use of ARIMA time series analysis methods, as well as steps for noise reduction and forecasting using Wavelet wavelet functions. Through these methods, the text provides sales forecast results for different vegetable categories for the coming week for the superstore management's reference.
... Many perennial streams originate from the Himalayas, but in recent times, because of climate change, the Himalayas have faced an environmental threat. In the Himalayan region, the rainfall pattern has changed, and the temperature of the region has also changed (Rawat et al. 2022). Water scarcity has increased in the Himalayan region, which has had an adverse impact on agriculture and horticulture. ...
This research article reflects the use of an integrated approach of geospatial and social sciences for the planning and management of micro watersheds in Dharampur Block, Himachal Pradesh. The micro-watershed area had many problems related to soil erosion, scarcity of water sources during the summer season, and unsustainable livelihood options. These problems were identified through the field survey and presented through the geospatial sciences. In the initial stage of watershed development, the plan was implemented to solve the problems of soil erosion, water scarcity through check dams and spring shed management, and the revival of natural water sources. The sustainable livelihood problem tries to be solved by uplifting agriculture and horticulture and linking people with microenterprises such as beekeeping, poultry, and dairy. The weaker sections and communities of the micro watershed were the targeted groups under the section of water social and economic development. The project implementation period starts in 2022 and ends in 2027, and its results shall be evaluated after the completion of this project.
... Using the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC), ARIMA (0,0,1) (1,1,2) 12 was found to be the best fit. Rawat et al. (2022) assessed the forecasting efficiency and prediction pattern of rainfall over the western Himalayas of India by the NAR and ARIMA models. The study concluded that Nonlinear autoregressive neural network (NARNN) models can also be used. ...
This study utilizes time series analysis and machine learning techniques to model and forecast rainfall patterns across different seasons in India. The statistical models, i.e., autoregressive integrated moving average (ARIMA) and state space model and machine learning models, i.e., Support Vector Machine, Artificial Neural Network and Random Forest Model were developed and their performance was compared against XGBoost, an advanced machine learning algorithm, using training and testing datasets. The results demonstrate the superior accuracy of XGBoost compared to the statistical models in capturing complex nonlinear rainfall patterns. While ARIMA models tend to overfit the training data, state space models prove more robust to outliers in the testing set. Diagnostic checks show the models adequately capture the time series properties. The analysis indicates essential unchanging rainfall patterns in India for 2023-2027, with implications for water resource management and climate-sensitive sectors like agriculture and power generation. Overall, the study highlights the efficacy of modern machine learning approaches like XGBoost for forecasting complex meteorological time series. The framework presented enables rigorous validation and selection of optimal techniques. Further applications of such sophisticated data analysis can significantly enhance planning and research on the Indian monsoons amidst climate change challenges.
... Classification is two step processes: learning or training step where data is analyzed by a classification algorithm and testing step where data is used for classification and to estimate the accuracy of the Classification [26][27][28]. The input to classifier models C4.5 decision tree, Support Vector Machine (SVM), and Naïve Bayes (NB) is important features that selected in previous step and the output of these models is classified class breast cancer malignant or benign tumor [29]. ...
Breast cancer is a significant public health concern worldwide, and early detection is crucial for its treatment. Although breast cancer has been extensively studied, there is still room for improvement in its classification accuracy. This study aims to improve the classification accuracy of breast cancer by applying information gain feature selection and machine learning techniques to the Wisconsin Diagnostic Breast Cancer (WDBC) dataset. The information gain method is utilized to reduce feature characteristics, and machine learning algorithms such as support vector machine (SVM), naive Bayes (NB), and C4.5 decision tree are employed for breast cancer classification. The study also conducts a comparison analysis based on accuracy value. The proposed model achieves maximum classification accuracy (100%) and a weighted average for precision (100%) and recall (100%) using a C4.5 decision tree, while SVM accuracy (98.42%) and weighted average for precision (98.17%) and recall (98.58%) are achieved using a C4.5 decision tree. The NB algorithm attains an accuracy of 96%, with a weighted average for precision (18.57%) and recall (50%). The proposed model's results are compared to similar studies and demonstrate significant progress, indicating new opportunities for breast cancer detection.
... Observing partial autocorrelation and autocorrelation assist in choosing p and q parameters. Statistical methods cannot identify nonlinear patterns and irregularities in time series [15]. ...
The city of Jeddah experienced a severe flood in 2020, resulting in loss of life and damage
to property. In such scenarios, a flood forecasting model can play a crucial role in predicting flood
events and minimizing their impact on communities. The proposed study aims to evaluate the
performance of machine learning algorithms in predicting floods and non-flood regions, including
Gradient Boosting, Extreme Gradient Boosting, AdaBoosting Gradient, Random Forest, and the Light
Gradient Boosting Machine, using the dataset from Jeddah City, Saudi Arabia. This study identified
fourteen continuous parameters and various classification variables to assess the correlation between
these variables and flooding incidents in the analyzed region. The performance of the proposed
algorithms was measured using classification matrices and regression matrices. The highest accuracy
(86%) was achieved by the Random Forest classifier, and the lowest error rate (0.06) was found with
the Gradient Boosting regressor machine. The performance of other algorithms was also exceptional
compared to existing literature. The results of the study suggest that the application of these machine
learning algorithms can significantly enhance flood prediction accuracy, enabling various industries
and sectors to make more informed decisions.
This chapter presents a comparative discussion of the temporal and spatial distribution of the seasonal snow cover of Ladakh. The monitoring of snow was performed, including ground truthing, using a multi-disciplinary approach, applying the power of advanced remote sensing techniques for the period 2000–2019. Ladakh marks the transitional area between predominantly receding glaciers of the Central Himalaya and some advancing glaciers of the Karakoram, in comparison with the adjoining South Asian Mountain Belt, and it features monsoonal and cold-arid regional climatic differences. Thus, the snow cover is essential to understand the land-atmosphere interaction, and it serves as a primary vital water resource to ensure the availability of water for the irrigation of agricultural fields and other applications.
The soil is well-known to offer and regulate a wide range of ecosystem services and play a significant role in supporting human life. The 17 Sustainable Development Goals (SDGs) adopted in the United Nations general assembly in September 2015 have also reinforced the importance of various soil functions, ecosystem services, carbon sequestration by soil, prevention of soil degradation, and their influence on life on earth and global climate change. Although soils are not clearly mentioned in the SDGs, a few SDGs are closely related to the characteristics and functions of soil. Among the SDGs defined and approved in general assembly of UN in 2015, eight directly or indirectly include soil functions.
The Indian Himalaya is an important region in terms of global significance due to its enriched ecological and sociocultural diversity and vast range of ecosystem services for the very existence of humanity. The better and improved soil ecosystem services provided by different land uses of this region directly feed into the realisation and achievement of the soil-related SDGs of the United Nations. However, the soil ecosystem services provided by Indian Himalaya are under remarkable pressure due to the rapid growth in human population, uncontrolled urbanisation, developmental activities, increase in agricultural activities, climate change, and the related changes in land use and land cover. Various studies conducted in this part of country indicated a general decline in forest cover area and an increasing trend both in fragmentation of forest cover and in the built-up areas, barren areas and croplands in recent years. This observation entails the degradation of ecosystems and the decrease in the potential for the sustainable flow of ecosystem services from this region. In view of these facts, this chapter aims to provide a systematic review of the soil ecosystem services provided by various land uses (from existing literature) and their ultimate sustainable management in order to achieve soil-related SDGs in Indian Himalayan ecosystem.
Accurate GDP forecasts are vital for strategic decision-making and effective macroeconomic policies. In this study, we propose an innovative approach for Chongqing's GDP prediction, combining the LASSO method with the CWOA—BP–ARIMA model. Through meticulous feature selection based on Pearson correlation and Lasso regression, we identify key economic indicators linked to Chongqing's GDP. These indicators serve as inputs for the optimized CWOA–BP–ARIMA model, demonstrating its superiority over Random Forest, MLP, GA–BP, and CWOA–BP models. The CWOA–BP–ARIMA model achieves a remarkable 95% reduction in MAE and a significant 94.2% reduction in RMSE compared to Random Forest. Furthermore, it shows substantial reductions of 80.6% in MAE and 77.8% in RMSE compared to MLP, along with considerable reductions of 77.3% in MAE and 75% in RMSE compared to GA–BP. Moreover, compared to its own CWOA–BP counterpart, the model attains an impressive 30.7% reduction in MAE and a 20.46% reduction in RMSE. These results underscore the model's predictive accuracy and robustness, establishing it as a reliable tool for economic planning and decision-making. Additionally, our study calculates GDP prediction intervals at different confidence levels, further enhancing forecasting accuracy. The research uncovers a close relationship between GDP and key indicators, providing valuable insights for policy formulation. Based on the predictions, Chongqing's GDP is projected to experience positive growth, reaching 298,880 thousand yuan in 2022, 322,990 thousand yuan in 2023, and 342,730 thousand yuan in 2024. These projections equip decision-makers with essential information to formulate effective policies aligned with economic trends. Overall, our study provides valuable knowledge and tools for strategic decision-making and macroeconomic policy formulation, showcasing the exceptional performance of the CWOA–BP–ARIMA model in GDP prediction.
The Indian meteorological department provides only short term forecasts of weather, but long term forecasting is required for effective planning of agro production related activities. Generally, long term rainfall forecasting can be achieved by two approaches, traditional almanac forecasting (TAF), and scientific weather forecasting (SWF). TAF is based on observations and experience using a combination of meteorological and astronomical indicators, activities of animals, insects, and plants, and almanacs (also called as panchangams). SWF is based on past records of climate in a particular region using mathematical models. The main objective of this study is to recommend crops based on forecasted rainfall in Tamilnadu, India using TAF and SWF models. In this work, a hybrid non-linear autoregression Neural Network is used as SWF and saint Kaikkadar Rainfall Prediction Methodology is used as TAF. An empirical study is conducted to compare the performance of TAF and SWF models It is clear that the hybrid AF NARNN model outperformed the baseline AF NARNN model in terms of RMSE value. The hybrid AF NARNN model improved forecast accuracy by 4.0%, 3.3%, 5.4%, 5.1%, 6.9%, 0.3%, 5.3%, 2.6%, 1.0%, 6.0, and 6.0, respectively, with the exception of October. The prediction of TAF model is more than 65% similar to SWF. As a result, the TAF model can be utilised as a tool for long-term forecasting of activities related to agriculture.KeywordsAlmanacPanchangamRainfall predictionNeural NetworkNARHybrid NARNNCrop recommendation
India is an agrarian society where 13.7% of GDP and 50% of workforce are involved with agriculture. Rainfall plays a vital role in irrigating the land and replenishing the rivers and underground water sources. Therefore the study of rainfall is vital to the economic development and wellbeing of the nation. Accurate prediction of rainfall leads to better agricultural planning, flood prevention and control. The seasonal artificial neural networks can predict monthly rainfall by exploiting the cyclical nature of the weather system. It is dependent on historical time series data and therefore independent of changes in the fundamental models of climate known collectively as manmade climate change. This paper presents the seasonal artificial neural networks applied on the prediction of monthly rainfall. The amounts of rainfall in the twelve months of a year are fed to the neural networks to predict the next twelve months. The gradient descent method is used for training the neural networks. Four performance measures viz. MSE, RMSE, MAD and MAPE are used to assess the system. Experimental results indicate that monthly rainfall patterns can be predicted accurately by seasonal neural networks.
Weather factors such as temperature and humidity are indispensable for good agriculture.The best-suitable products can be selected according to the optimal of these weather factors. In this study, data on maximum temperature, minimum temperature, morning relative humidity and evening relative humidity was analyzed from 31 st January, 1921 to 31 st December, 2020 in India. The BATS (Exponential Smoothing Method + Box-Cox Transformation + ARMA model for residuals) and TBAT (BATS + Trigonometric Seasonal) models are conducted for forecasting procedures. According to some selection criteria, the best models are specified for all weather factors. Extensive tables and graphics are presented in the study. The data series was divided into train set and test set. The result obtained from train set based on goodness of fit, BATS model performed as a best model. For error estimation using testing data set, BATS model performed well for maximum temperature, morning and evening relative humidity. Both models were performed significantly atpar in minimum temperature data series. Using the results and forecasts obtained in this study, the researcher or scientist should be focused on the weather condition which is more concerned parameter for agriculture.
In this study, 9 remaining water components of Madian River in the Baraftab were considered. Autoregressive Integrated Moving Average modeling techniques are considered appropriate in creating and predicting components. SO42–, Na + and SAR show a declining fashion despite the different properties of brilliant water, which show an increasing lightness. However, practicing a delay to the position of the style desk was sure that the time group was ready for painting. Similarly, the time group study suggests that there is an unfamiliar location increase mode for all components except SO42–, Na + , and SAR. EC, Cl–, Ca2 + , Mg2 + and HCO3– show an increasing style, which is suggested to reduce the prominent water within the area. Depend on the sphere revisions, the extreme successful and relative population density, rising the use of artificial frameworks, departure city effluents and mainstream of rural manure in conservative method via streams, inopportune plans of burial litters, dispersal of nonsenses and disorders in floor waters and watercourses which later influx via streams are taken into deliberation because the important reasons of water brilliant worsening. Agronomic effluents and livestock are different causes that pollute floor water. Also, the water hazard outstanding worsening is rising because of extreme public boom with inside the area and green changes are energetic with inside the zone to exclude you larger ecological devastation.
Thailand is one of the countries where foot and mouth disease outbreaks have resulted in
considerable economic losses. Forecasting is an important warning technique that can allow authorities to establish an FMD surveillance and control program. This study aimed to model and forecast the monthly number of FMD outbreak episodes (n-FMD episodes) in Thailand using the time-series methods, including seasonal autoregressive integrated moving average (SARIMA), error trend seasonality (ETS), neural network autoregression (NNAR), and Trigonometric Exponential smoothing state–space model with Box–Cox transformation, ARMA errors, Trend and Seasonal components (TBATS), and hybrid methods. These methods were applied to monthly n-FMD episodes (n = 1209) from January 2010 to December 2020. Results showed that the n-FMD episodes had a stable trend from 2010 to 2020, but they appeared to increase from 2014 to 2020. The outbreak episodes followed a seasonal pattern, with a predominant peak occurring from September to November annually. The single technique methods yielded the best-fitting time-series models, including SARIMA(1, 0, 1)(0, 1, 1)12,NNAR(3, 1, 2)12, ETS(A, N, A), and TBATS(1, f0, 0g, 0.8, f< 12, 5 >g. Moreover, SARIMA-NNARand NNAR-TBATS were the hybrid models that performed the best on the validation datasets. Themodels that incorporate seasonality and a non-linear trend performed better than others. The forecastshighlighted the rising trend of n-FMD episodes in Thailand, which shares borders with severalFMD endemic countries in which cross-border trading of cattle is found common. Thus, controlstrategies and effective measures to prevent FMD outbreaks should be strengthened not only inThailand but also in neighboring countries.
Sugarcane industry is of crucial importance to the South Asian countries. These countries depend heavily on agriculture and the sugarcane industry has immense potential to contribute towards its economic development. Hence, the precise and timely forecast of sugarcane production is of concern for farmers, policy makers and other stakeholders. In this manuscript, we strived to forecast the production and growth rate of this important commodity using standard statistical approaches. The ARIMA (Auto Regressive Integrated Moving Average) and ETS (Exponential Smoothing) models were applied and compared on the basis of their forecasting efficiency for South Asia countries. This study also investigated the trends in sugarcane production in the region and studies the causes of the decline in production of sugarcane in Sri Lanka and Bangladesh. Furthermore, the expected production for following 7 years was computed using both models. In addition, we also calculated the projected growth rates of sugarcane production of South Asian countries over the years 2020-2027.
In different parts of the globe, accurate rainfall forecasting models are needed to forecast rainfall in real-time during the typhoon season to prevent disasters caused by heavy rainfall in the region. The current study developed a time series based analysis technique to predict precipitation during the monsoon season. As evidenced by the investigations in this study, a data-driven technique has considerable scope for predicting the future variables and patterns from existing data, mainly when applied to complex and challenging natural phenomena such as rainfall. The ANN-based technique has an immense potential to predict rainfall data using lagged time series analysis model. We have investigated rainfall patterns, their variability in Maharashtra state and future rainfall prediction through the present study. The neural network auto-regression model put forth is a promising technique for rainfall prediction. The model's performance is evaluated concerning error rate and model fit and exhibits reasonably good performance.
Accurate rainfall forecasting in watersheds is of indispensable importance for predicting streamflow and flash floods. This paper investigates the accuracy of several forecasting technologies based on Wavelet Packet Decomposition (WPD) in monthly rainfall forecasting. First, WPD decomposes the observed monthly rainfall data into several subcomponents. Then, three data-based models, namely Back-propagation Neural Network (BPNN) model, group method of data handing (GMDH) model, and autoregressive integrated moving average (ARIMA) model, are utilized to complete the prediction of the decomposed monthly rainfall series, respectively. Finally, the ensemble prediction result of the model is formulated by summing the outputs of all submodules. Meanwhile, these six models are employed for benchmark comparison to study the prediction performance of these conjunction methods, which are BPNN, WPD-BPNN, GMDH, WPD-GMDH, ARIMA, and WPD-ARIMA models. The paper takes monthly data from Luoning and Zuoyu stations in Luoyang city of China as the case study. The performance of these conjunction methods is tested by four quantitative indexes. Results show that WPD can efficiently improve the forecasting accuracy and the proposed WPD-BPNN model can achieve better prediction results. It is concluded that the hybrid forecast model is a very efficient tool to improve the accuracy of mid- and long-term rainfall forecasting.
The estimation of rainfall is one of the most critical and daunting challenges in today’s environment. Weather and rainfall are typically extremely nonlinear and dynamic, needing sophisticated machine models and simulation for forecasting accurately. The economy of India is agriculture and is focused primarily on crop production and precipitation. Predictions of rainfall are important for all farmers to assess crop productivity. Rainfall forecast involves the application of science and technology to determine weather conditions. In order to utilize water supplies efficiently, the crop productivity and the pre-program of water systems, it is necessary to determine the precipitation in detail. The actions of such nonlinear processes can be modeled using an Artificial Neural Network (ANN). Most researchers in this area have been effectively utilizing ANN for the past 25 years. This article offers you an summary of some of the methodologies valid for using ANN for rainfall prediction by numerous researchers.The survey also states that forecasts of rainfall using ANN technologies are more accurate than conventional mathematical and numerical approaches.
A random variable can take very large or very small values known as extreme values. In some instances, the researcher’s interest lies mainly on these extreme values like maximum (or minimum) temperature, maximum (or minimum) amount of precipitations, maximum level of flood water, maximum (or minimum) wind speed, maximum (or minimum) level of disease, or pest infestation in a particular crop or season. These types of extreme values can be modelled by generalized extreme value (GEV) distribution, theorized by McFadden in 1978. The type I GEV distribution, i.e. Gumbel distribution, has been extensively studied by many authors, but due to its constant skewness and kurtosis, it has limited practical application. To this end, different generalizations of Gumbel distribution have been proposed in different pieces of literature. In this manuscript, generalized Gumbel (GG) distribution is obtained from generalized multivariate Gumbel (GMVGB) distribution specified in Demirhan and Hamurkaroglu (Journal of Statistical Planning and Inference 141(Adeyemi and Ojo 2003):1141-1152, 2011) and different properties of this distribution have been derived. The continuous increasing of atmospheric concentration of greenhouse gases, mainly due to human activity, has led to global climate change. Climate change can in turn modify the occurrence and intensity of extreme weather and climate events—heat waves and sea-level rise. In the present study, one such climatic variable, i.e. the annual maximum temperature data of India, has been taken for the period of 1901 to 2017 and various extreme value distributions, namely GEV, Gumbel and GG, have been fitted. The obtained results indicate that the GG distribution is better fitted than the other two competing distributions.
ABSTRACT
Background: The goal of this study was to forecast pulse production in six countries: Afghanistan, Bangladesh, China, India, Nepal and Pakistan (2020-2027). In this study, time series forecasting was used.
Methods: The data series were divided into training set from 1961 to 2015 for model building, testing set from 2016 to 2019 for validation and finally, after selecting the best model, forecast was used from 2020 to 2027, the models were compared. The best-fit
model was chosen based on the minimum ME, RMSE, MAE, MPE, MAPE, MASE, ACF1 values on the training data set and the MAPE values on the testing data set.
Result: The best fitted model for India was NNAR (1,1). Similar to Afghanistan, the best fit model for forecasting was NNAR (3,2). The best fit model for forecasting in China was ARIMA (0,1,1). The best fit model for forecasting in Nepal was ARIMA (1,1,0). The best fit
model for forecasting in Pakistan was ETS (A, N, N) (M, N, N). With a 15.73 per cent growth rate from 2020 to 2027, the best models predict that the production of pulses in (Afghanistan, China, India) will increase until 2027. India will continue to be the largest
producer of pulses among the six countries, with production expected to reach 1088.778 thousand tons in 2027. Afghanistan and China have extreme growth rates of 25.19% and 11.95%, respectively, while the rest of the countries have relatively stable production
volumes. These results may be crucial for developing an effective agriculture production policy, whether by providing forecasted production values or evaluating such policies