The Statistical Techniques of Artificial Intelligence with python For Time Series Forecasting

The Statistical Techniques of Artificial Intelligence with python For Time Series Forecasting | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article The Statistical Techniques of Artificial Intelligence with python For Time Series Forecasting SEKHAR PRANITHA This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6373055/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract The Artificial Intelligence with the Time series forecasting with a critical component which enables the organization to make a structured decisions on the historic of the data. This paper explores the several techniques of statistics for the time series forecasting which includes Autoregressive Integrated Moving Average(ARIMA), Exponential Smoothing State Space Model(ETS), and Seasonal Decomposition of Time Series(STL). We can also implement these methods to evaluate the performance of the techniques with the dataset of Air passengers from 1949 to 1960. The ARIMA which captures the underlying patterns of the data, whereas ETS demonstrates the accuracy of the data with the seasonal components, and STL decomposes the technique which provides valuable things according to the trend of the time series. This study not only focuses on the strength and the weakness of the methods but also provides practical implementation with the code which will be increasing in the knowledge in time series forecasting within the field of Artificial Intelligence. The main aim of this paper to compare these techniques by analyzing the results of the study which can be applied to the Artificial Intelligence and providing which will be most suitable approach for the forecasting of time series needs. Autoregressive Integrated Moving Average(ARIMA) Exponential Smoothing State Space Model(ETS) Seasonal Decomposition of Time Series(STL) Artificial Intelligence Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction The Time Series Forecasting which is an extremely analytical technique for the predicting of the future values based on the history of the data which has been collected over the time. It is applied in various domains which includes the finance, economics, healthcare and managements where the time of the patterns influences the decision-making process. As it is increasing the strategies with the ability to accurate the future trends of the data which will forecast the data. The Artificial Intelligence in the time series forecasting plays an important role for the predictive analytics. The historical data which can detect the patterns and make prediction of the future trends. The forecasting can make a better improvement in strategies in different industries where it can improve its outcome through the time. There are different techniques which has been used or developed for the time series forecasting by the ARIMA which is mostly used methods for its flexibility in different models for the time series data. This can combine with the autoregression and averages to capture it efficiency effectively. When it needs more robust which will combine with Seasonal and form it as SARIMA. The ETS models which utilize for the smoothing for capturing the components of the time series which make the forecasting more effective. Which will be used for the trends, Errors, and Seasonality of the functions. The STL method which decomposes the separate time series of the seasonal trend and tits components. It is mainly used the seasonal components, Trend Component and the Residual Component which will robust the outliers. Literature Review [ 1 ] This paper shows that how the Machine learning and deep learning was happened to be in the form with some of the python libraries with the traditional statistical methods. [ 2 ] This article has explained the concepts with the transformation of the ARIMA and the LSTM which will be data- driven concepts where the forecasting of the data with the integrating of the database. [ 3 ] According to this journal the forecasting of the data is done by the ARIMA which were all the inputs of real time with the call of the action in which the data is consider according to its scalability efficiency and resilience [ 4 ] In this article the stock exchanges of the data done with the ARIMA model which is primarily used for its accuracy which may predict according to its standards. [ 5 ]according to this article the method of doing with the algorithm ARIMA and the extension of its SARIMA it may vary its range and its factor. Although the dataset result may not be significant but they won’t reject the Null Hypothesis. Methodology 4.1 Data Collection The Air Passenger dataset from 1949 to 1960, with the repository. The dataset which has only two columns with date (month/year) and the number of passengers and then load the data of air passenger in google Collab/notebook 4.2 Data Preprocessing Outliers and Missing values: To check the data whether they are any missing values or the outliers. Visualization: The data which is visualized to identify its trends and the patterns seasonally. Checking Stationarity: The time series which was assessed by using the Augmented Dickey-Fuller test. If it is not found then the non-stationary difference will stabilize its mean and according to it will follow the series. 4.3 STL Decomposition The Decomposition of the STL applied to separate the series of the time into the seasonal, trends and the components of the residual. This step may aid into the understanding of the structure which is underlying of the data and it also prepares its future modelling. 4.4 Selection and Implementation Model 4.4.1 ETS Model The Exponential Smoothing State Space Model which was fitted to the data. This model parameters were selected on the basis of its characteristics of the dataset, which will be considering the trend and the components seasonal. 4.4.2 ARIMA Model This model will implement following with the identification of its parameters for the user of the Autocorrelation Function (ACF) and the Partial Autocorrection Function (PACF) with the plots. 4.5 Forecasting The forecasting was generated for the next coming months by using the both ETS and ARIMA models. The forecasting was stored in a data frame for comparison. 4.6 Model Evaluation The metrics of the forecasting models was evaluated using the metrics such as Mean Squared Error (MSE) and Mean Absolute Error (MAE). The actual values for the forecast time were compared against the predicted values of the assess of the accuracy. Results The result indicated the Exponential Smoothing State model which was performed out both the ARIMA and STL models which the terms of both Absolute Error and Squared Error. The ETS model which has been achieved significantly lower MAE of 31.79 and MSE of 1541.40, which will demonstrate the effective for capturing the below trends and patterns of the seasonal according to the Air Passenger Dataset. In the contrast, the ARIMA model with the highest error metrics, with an MAE of 85.25 and an MSE of 9506.18, which will be indicating its struggling accurate forecast of the passenger numbers. And the STL model performs, with an MAE of 6808 and MSE of 7250.11, but still it has the short of the accuracy with the achieved by the ETS Model. The Visualization of comparison of actual and forecasted values for the model with the further findings, with which the ETS model aligning the actual passenger count, while the ARIMA model shows the greater deviations. Future Directions The some of the future directions with the field of time series forecasting: The integration of the models with the algorithms of the machine learning like Random Forests, Gradient Boosting or Neural Networks with the statistical models like ARIMA, STL, and ETS. Implementing the methods which the combination of the multiple models which could lead to improved forecasting metrics. The implementation of the real-time forecasting system with the continues update of the prediction as the new data will be available which could be beneficial for the industries for the decision making for the crucial methods The future study which would investigate the extra metrics beyond the MAE and MSE such as MAPE, RMSE and others for the understanding of the performance of model. These future methods which can continue for enhancing the field of the time series forecasting which will lead more accurate and reliable across the various applications Conclusion This paper conducted as a comparative analysis for the three-time series forecasting techniques that are ARIMA, STL and ETS for the dataset of Air Passengers. The result demonstrates the model more significantly for the ETS for the outperformed both the ARIMA and STL in the terms of forecasting accuracy which has been evidenced by its values of MAE and MSE for less values. These findings suggest the dataset which exhibits has a clear trend with the ETS model may be the preferred choice for the forecasting. Further for research could explore for the machine learning techniques with the statistical models to enhance the accuracy. However by applying these models with the other datasets for the different domains which may provide large valuable sets for the effectiveness References Modern Time series forecasting with python 2nd Edition Forecasting Techniques in predictive Analytics leveraging database management for scalability and the real time insights Techniques of forecastiong with valuable insights according to the real time applications Study of Effective of the series modelling in forecasting the stock prices The effective model with the series according to the ARIMA for the time series forecasting Time Series Analysis: Forecasting and Control - Jenkins, G. M (1970) Automatic Time Series Forecasting: The forecast package for R. Journal of Statistical Software - Khandakar, Y (2008) The Analysis of Time Series: An Introduction (6th ed.) - Chatfield, C (2004) Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6373055","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":438226281,"identity":"b4ca7f90-3895-433f-b9d8-863dac8d788b","order_by":0,"name":"SEKHAR PRANITHA","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCElEQVRIiWNgGAWjYLCCBIYDQJKNDUjYyIEEDjwgqCUBriXNGKwlgbA1cC2HExsg9uIG5tKHnz14+OOOvHx7W9qDn3uY0+eHHX4ItMVOTrcBuxbLvjRzg4SEZ4Ybzhw7btjzjC134+00A6CWZGOzA9i1GJxhMJNISDjMuEEivU2C5wBP7sbZCSAtBxK34dTC/g2kxX7+/Odtkn8OSKQbzk7/QEALD9iWxIYbbMekeQ4YJMhL5+C3xbKHp0wiIe1w8oYzaWnSMgcSDDdI5xQcSDDA7RdzHvZtkj9sDtvObz9mJvnmwH95+dnpmz98qLCTw+l9TJED2MXxaJFvwK16FIyCUTAKRiYAAOL7ZwSQWZ2GAAAAAElFTkSuQmCC","orcid":"","institution":"ravindra college of engineering for women","correspondingAuthor":true,"prefix":"","firstName":"SEKHAR","middleName":"","lastName":"PRANITHA","suffix":""}],"badges":[],"createdAt":"2025-04-04 04:03:17","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":true,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":true},"doi":"10.21203/rs.3.rs-6373055/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6373055/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":80017894,"identity":"26b42417-d1fb-45ad-9f35-cfb659c4fdd3","added_by":"auto","created_at":"2025-04-07 03:59:36","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":68702,"visible":true,"origin":"","legend":"\u003cp\u003eThe Exponential Smoothing State Space Model which was fitted to the data. This model parameters \u0026nbsp;were selected on the basis of its characteristics of the dataset, which will be considering the trend and the components seasonal.\u003c/p\u003e","description":"","filename":"Picture1.png","url":"https://assets-eu.researchsquare.com/files/rs-6373055/v1/fdefaf246985045b27a913c6.png"},{"id":80017895,"identity":"4d043a6f-b8a9-4874-b600-7d3110d5f910","added_by":"auto","created_at":"2025-04-07 03:59:36","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":63487,"visible":true,"origin":"","legend":"\u003cp\u003eThe Decomposition of the STL applied to separate the series of the time into the seasonal, trends and the components of the residual. This step may aid into the understanding of the structure which is underlying of the data and it also prepares its future modelling.\u003c/p\u003e","description":"","filename":"Picture2.png","url":"https://assets-eu.researchsquare.com/files/rs-6373055/v1/61941eeab5d9ad7d1fb1e0f9.png"},{"id":80017898,"identity":"910566a3-105e-48ee-bf9b-55ff6523d243","added_by":"auto","created_at":"2025-04-07 03:59:36","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":42643,"visible":true,"origin":"","legend":"\u003cp\u003eThis model will implement following with the identification of its parameters for the user of the Autocorrelation Function (ACF) and the Partial Autocorrection Function (PACF) with the plots.\u003c/p\u003e","description":"","filename":"Picture3.png","url":"https://assets-eu.researchsquare.com/files/rs-6373055/v1/4bdea22b451a1487066bae6e.png"},{"id":80017901,"identity":"5f60e61a-1df6-4376-bf79-ebe00c60c706","added_by":"auto","created_at":"2025-04-07 03:59:37","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":36897,"visible":true,"origin":"","legend":"\u003cp\u003eUnnumbered image in the Results section.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-6373055/v1/78e35c0622f0d1e34023e3dc.png"},{"id":80018726,"identity":"a237c1b0-5d2b-4cba-a539-34403a515621","added_by":"auto","created_at":"2025-04-07 04:15:41","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":522173,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6373055/v1/69e55e70-9551-45fb-9fb3-1b0e906baf9e.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eThe Statistical Techniques of Artificial Intelligence with python For Time Series Forecasting\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe Time Series Forecasting which is an extremely analytical technique for the predicting of the future values based on the history of the data which has been collected over the time. It is applied in various domains which includes the finance, economics, healthcare and managements where the time of the patterns influences the decision-making process. As it is increasing the strategies with the ability to accurate the future trends of the data which will forecast the data.\u003c/p\u003e \u003cp\u003eThe Artificial Intelligence in the time series forecasting plays an important role for the predictive analytics. The historical data which can detect the patterns and make prediction of the future trends. The forecasting can make a better improvement in strategies in different industries where it can improve its outcome through the time.\u003c/p\u003e \u003cp\u003eThere are different techniques which has been used or developed for the time series forecasting by the ARIMA which is mostly used methods for its flexibility in different models for the time series data. This can combine with the autoregression and averages to capture it efficiency effectively. When it needs more robust which will combine with Seasonal and form it as SARIMA.\u003c/p\u003e \u003cp\u003eThe ETS models which utilize for the smoothing for capturing the components of the time series which make the forecasting more effective. Which will be used for the trends, Errors, and Seasonality of the functions.\u003c/p\u003e \u003cp\u003eThe STL method which decomposes the separate time series of the seasonal trend and tits components. It is mainly used the seasonal components, Trend Component and the Residual Component which will robust the outliers.\u003c/p\u003e"},{"header":"Literature Review","content":"\u003cp\u003e[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e] This paper shows that how the Machine learning and deep learning was happened to be in the form with some of the python libraries with the traditional statistical methods.\u003c/p\u003e \u003cp\u003e[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e] This article has explained the concepts with the transformation of the ARIMA and the LSTM which will be data- driven concepts where the forecasting of the data with the integrating of the database.\u003c/p\u003e \u003cp\u003e[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e] According to this journal the forecasting of the data is done by the ARIMA which were all the inputs of real time with the call of the action in which the data is consider according to its scalability efficiency and resilience\u003c/p\u003e \u003cp\u003e[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e] In this article the stock exchanges of the data done with the ARIMA model which is primarily used for its accuracy which may predict according to its standards.\u003c/p\u003e \u003cp\u003e[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]according to this article the method of doing with the algorithm ARIMA and the extension of its SARIMA it may vary its range and its factor. Although the dataset result may not be significant but they won\u0026rsquo;t reject the Null Hypothesis.\u003c/p\u003e"},{"header":"Methodology","content":"\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e4.1 Data Collection\u003c/h2\u003e \u003cp\u003eThe Air Passenger dataset from 1949 to 1960, with the repository. The dataset which has only two columns with date (month/year) and the number of passengers and then load the data of air passenger in google Collab/notebook\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Data Preprocessing\u003c/h2\u003e \u003cp\u003eOutliers and Missing values: To check the data whether they are any missing values or the outliers.\u003c/p\u003e \u003cp\u003eVisualization: The data which is visualized to identify its trends and the patterns seasonally.\u003c/p\u003e \u003cp\u003eChecking Stationarity: The time series which was assessed by using the Augmented Dickey-Fuller test. If it is not found then the non-stationary difference will stabilize its mean and according to it will follow the series.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e4.3 STL Decomposition\u003c/h2\u003e \u003cp\u003eThe Decomposition of the STL applied to separate the series of the time into the seasonal, trends and the components of the residual. This step may aid into the understanding of the structure which is underlying of the data and it also prepares its future modelling.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e4.4 Selection and Implementation Model\u003c/h2\u003e \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e \u003ch2\u003e4.4.1 ETS Model\u003c/h2\u003e \u003cp\u003eThe Exponential Smoothing State Space Model which was fitted to the data. This model parameters were selected on the basis of its characteristics of the dataset, which will be considering the trend and the components seasonal.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003e4.4.2 ARIMA Model\u003c/h2\u003e \u003cp\u003eThis model will implement following with the identification of its parameters for the user of the Autocorrelation Function (ACF) and the Partial Autocorrection Function (PACF) with the plots.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e4.5 Forecasting\u003c/h2\u003e \u003cp\u003eThe forecasting was generated for the next coming months by using the both ETS and ARIMA models. The forecasting was stored in a data frame for comparison.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e4.6 Model Evaluation\u003c/h2\u003e \u003cp\u003eThe metrics of the forecasting models was evaluated using the metrics such as Mean Squared Error (MSE) and Mean Absolute Error (MAE). The actual values for the forecast time were compared against the predicted values of the assess of the accuracy.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003eThe result indicated the Exponential Smoothing State model which was performed out both the ARIMA and STL models which the terms of both Absolute Error and Squared Error. The ETS model which has been achieved significantly lower MAE of 31.79 and MSE of 1541.40, which will demonstrate the effective for capturing the below trends and patterns of the seasonal according to the Air Passenger Dataset.\u003c/p\u003e \u003cp\u003eIn the contrast, the ARIMA model with the highest error metrics, with an MAE of 85.25 and an MSE of 9506.18, which will be indicating its struggling accurate forecast of the passenger numbers. And the STL model performs, with an MAE of 6808 and MSE of 7250.11, but still it has the short of the accuracy with the achieved by the ETS Model.\u003c/p\u003e \u003cp\u003eThe Visualization of comparison of actual and forecasted values for the model with the further findings, with which the ETS model aligning the actual passenger count, while the ARIMA model shows the greater deviations.\u003c/p\u003e"},{"header":"Future Directions","content":"\u003cp\u003eThe some of the future directions with the field of time series forecasting:\u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eThe integration of the models with the algorithms of the machine learning like Random Forests, Gradient Boosting or Neural Networks with the statistical models like ARIMA, STL, and ETS.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eImplementing the methods which the combination of the multiple models which could lead to improved forecasting metrics.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eThe implementation of the real-time forecasting system with the continues update of the prediction as the new data will be available which could be beneficial for the industries for the decision making for the crucial methods\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eThe future study which would investigate the extra metrics beyond the MAE and MSE such as MAPE, RMSE and others for the understanding of the performance of model.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003eThese future methods which can continue for enhancing the field of the time series forecasting which will lead more accurate and reliable across the various applications\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis paper conducted as a comparative analysis for the three-time series forecasting techniques that are ARIMA, STL and ETS for the dataset of Air Passengers. The result demonstrates the model more significantly for the ETS for the outperformed both the ARIMA and STL in the terms of forecasting accuracy which has been evidenced by its values of MAE and MSE for less values.\u003c/p\u003e \u003cp\u003eThese findings suggest the dataset which exhibits has a clear trend with the ETS model may be the preferred choice for the forecasting. Further for research could explore for the machine learning techniques with the statistical models to enhance the accuracy. However by applying these models with the other datasets for the different domains which may provide large valuable sets for the effectiveness\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eModern Time series forecasting with python 2nd Edition\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eForecasting Techniques in predictive Analytics leveraging database management for scalability and the real time insights\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTechniques of forecastiong with valuable insights according to the real time applications\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStudy of Effective of the series modelling in forecasting the stock prices\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThe effective model with the series according to the ARIMA for the time series forecasting\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTime Series Analysis: Forecasting and Control - Jenkins, G. M (1970)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAutomatic Time Series Forecasting: The forecast package for R. Journal of Statistical Software - Khandakar, Y (2008)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThe Analysis of Time Series: An Introduction (6th ed.) - Chatfield, C (2004)\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"ravindra college of engineering for woman","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Autoregressive Integrated Moving Average(ARIMA), Exponential Smoothing State Space Model(ETS), Seasonal Decomposition of Time Series(STL), Artificial Intelligence","lastPublishedDoi":"10.21203/rs.3.rs-6373055/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6373055/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe Artificial Intelligence with the Time series forecasting with a critical component which enables the organization to make a structured decisions on the historic of the data. This paper explores the several techniques of statistics for the time series forecasting which includes Autoregressive Integrated Moving Average(ARIMA), Exponential Smoothing State Space Model(ETS), and Seasonal Decomposition of Time Series(STL). We can also implement these methods to evaluate the performance of the techniques with the dataset of Air passengers from 1949 to 1960. The ARIMA which captures the underlying patterns of the data, whereas ETS demonstrates the accuracy of the data with the seasonal components, and STL decomposes the technique which provides valuable things according to the trend of the time series. This study not only focuses on the strength and the weakness of the methods but also provides practical implementation with the code which will be increasing in the knowledge in time series forecasting within the field of Artificial Intelligence. The main aim of this paper to compare these techniques by analyzing the results of the study which can be applied to the Artificial Intelligence and providing which will be most suitable approach for the forecasting of time series needs.\u003c/p\u003e","manuscriptTitle":"The Statistical Techniques of Artificial Intelligence with python For Time Series Forecasting","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-07 03:59:32","doi":"10.21203/rs.3.rs-6373055/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"39c409ba-2102-4bf3-acaf-2dc24ba290c9","owner":[],"postedDate":"April 7th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-04-07T03:59:32+00:00","versionOfRecord":[],"versionCreatedAt":"2025-04-07 03:59:32","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6373055","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6373055","identity":"rs-6373055","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}