Using the least squares method to find predicted values
Journal Title: Modern Innovations, Systems and Technologies - Year 2024, Vol 4, Issue 4
Abstract
The least squares method is used to analyze experimentally obtained data. The purpose of this method is to find a function that is closest to the results obtained in practice. This method makes sense to use when constructing a short-term forecast. Thus, it is necessary to recalculate the theoretically obtained values when new data arrives. The advantages of the considered method include the relative simplicity of implementation, its efficiency and versatility. In other words, this method can be used in various problems in which it is necessary to obtain a preliminary forecast by which it is possible to evaluate the system under consideration. In the course of the work, formulas were derived for finding the parameters that are used to calculate theoretical values. Their derivation consists in minimizing the linear regression function, which is produced by equating its partial derivatives to zero. However, it is worth noting that the least squares method allows you to find functions only for linearly dependent quantities. The least squares method is not used to find nonlinear curves. Moreover, before using the least squares method, it is necessary to check the original value for outliers. Outliers are understood as values that differ significantly from the total value.
Authors and Affiliations
E. A. Lykova, E. S. Lykova
Keywords
Related Articles
- EP ID EP755372
- DOI 10.47813/2782-2818-2024-4-4-0213-0222
- Views 27
- Downloads 0
Analysis of thyroid nodule ultrasound images by image feature extraction technique
The most frequent left thyroid nodule is the presence of thyroid nodules that have never been seen before. With X-ray computed tomography (CT) being used more often in diagnosing thyroid disorders, however, image process...
Structure and properties of sintered aluminum-based alloys produced by extrusion
This article examines the structure and properties of extruded sintered aluminum-based alloys. Two methods for producing alloys have been studied: activated liquid-phase sintering and intensive hot processing. A comparis...
Features of wave propagation in viscous media in pipes
This article presents research on wave propagation, specifically focusing on their propagation in half-spaces, at the contact between two bodies, as well as in various plate-like and cylindrical forms. In the region of l...
Comparative analysis of C++ and Java programming languages from the point of view of code safety assurance
A programming language is a set of commands that help computers, cell phones, tablets, and other electronic devices function as intended and perform various tasks. There are many types of programming languages, and choos...
Application of the linear regression method to predict vehicle fuel consumption
This article discusses the use of machine learning to predict vehicle fuel consumption. Linear regression is used for this task. The stages of data preparation are described, including data processing and elimination of...