The Effect of a Transformation Group on the Quality Indicator of a Linear Regression Model
УДК 514.172; 519.654
Abstract
The construction of functional dependencies between the observed phenomena is an important area of modern applied mathematics. The basis of such constructions is often a statistical data array. The adequacy of the models obtained directly depends on the quality of these data. In general, one has to choose one of the possible models, based on a certain indicator. However, the resulting samples may be in some sense identical, but the models constructed will be different.
This paper discusses one of the methods for constructing linear regression — the method of least squares. The problem of changing the quality functional of a regression model under the orthogonal transformation of the initial data set is studied. A geometric interpretation of the regression model itself and its functional quality, as well as the statistical indicator of the relationship between variables — the correlation coefficient, is given. Formulas are shown explicitly showing the relationship between the functionals of quality during rotation of a set relative to one of the axes of coordinates in two- and threedimensional spaces. Based on the formulas obtained, an algorithm is presented that allows one to obtain the value of the quality functional with any proper movement of n-dimensional space.
Downloads
Metrics
References
Greene W.H. Econometric Analysis. N.Y., 2008.
Дрейпер Н, Смит Г. Прикладной регрессионный анализ. Множественная регрессия // Applied Regression Analysis. М., 2007.
Стрижов В.В., Крымова Е.А. Методы выбора регрессионных моделей. М., 2010.
Amiri-Simkooei A.R., Jazaeri S. Weighted total least squares formulated by standard least squares theory // Journal of Geodetic Science. 2012. V. 2(2).
Пономарев И.В., Славский В.В. О геометрической интерпретации метода наименьших квадратов // Известия Алт. гос. ун-та. 2012. № 1-1(73).
Пономарев И.В. Геометрические преобразования модели линейной регрессии // Труды семинара по геометрии и математическому моделированию. 2018. № 4.
Лагутин М.Б. Наглядная математическая статистика. В 2-х т. М., 2003.
Кендалл М., Стюарт А. Статистические выводы и связи. М., 1973. Т. 2.
Берже М. Геометрия. М., 1984. Т. 1.
Шафаревич И.Р., Ремизов А.О. Линейная алгебра и геометрия. М., 2009.
Izvestiya of Altai State University is a golden publisher, as we allow self-archiving, but most importantly we are fully transparent about your rights.
Authors may present and discuss their findings ahead of publication: at biological or scientific conferences, on preprint servers, in public databases, and in blogs, wikis, tweets, and other informal communication channels.
Izvestiya of Altai State University allows authors to deposit manuscripts (currently under review or those for intended submission to Izvestiya of Altai State University) in non-commercial, pre-print servers such as ArXiv.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License (CC BY 4.0) that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).