Show simple item record

dc.contributor.advisorLin, Runchang
dc.creatorDe La Cruz Hernandez, Jacinto N/A
dc.date.accessioned2020-04-29T13:29:54Z
dc.date.available2020-04-29T13:29:54Z
dc.date.created2019-12
dc.date.submittedDecember 2019
dc.identifier.urihttps://hdl.handle.net/2152.4/230
dc.description.abstractResearchers in many fields share a deep interest in the sunspot activity of the Sun. This kind of solar activity has many consequences for human interests, and thus, it is important to study the Sun’s behavior and be able to predict future sunspot appearances. This task is a problem of time series forecasting. Some of the most popular methods used to forecast time series data are autoregressive models that predict future data points using a linear combination of previous values. Then, there are neural networks, a popular new tool for regression that can perform with outstanding results. These machine learning models have been shown to forecast time series successfully. In this thesis we use a variety of neural network architectures based on the classic AR models to predict future values of sunspot activity.
dc.format.mimetypeapplication/pdf
dc.subjectSunspots, Forecasting, Autoregressive
dc.titleForecasting Sunspot Number Time Series with Autoregressive Models
dc.typeThesis
dc.date.updated2020-04-29T13:29:55Z
dc.type.materialtext
thesis.degree.nameMaster of Science
thesis.degree.levelMasters
thesis.degree.disciplineMathematics
thesis.degree.grantorTexas A&M International University
thesis.degree.departmentMathematics & Physics


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record