Non-symmetric matrix and unified theory of least squares
Shiqing Wang, Ying Ma, Zhijun Feng
Abstract
Shiqing Wang, Ying Ma, Zhijun Feng
Abstract
Unified theory of least squares was given by Rao (197l, 1972, 1973) and Rao and Mitra (197l). In the generalized inverse (g-inverse) matrix (R + XGX')− involved in unified theory of least squares, G is any symmetric matrix. We give the properties with matrix quadratic form of non-symmetric matrix, and extend symmetric matrix G to an arbitrary square matrix. i.e. G is not necessarily symmetric only if matrix XGX' is symmetric.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Unified theory of least squares was given by Rao (197l, 1972, 1973) and Rao and Mitra (197l). In the generalized inverse (g-inverse) matrix (R + XGX')− involved in unified theory of least squares, G is any symmetric matrix. We give the properties with matrix quadratic form of non-symmetric matrix, and extend symmetric matrix G to an arbitrary square matrix. i.e. G is not necessarily symmetric only if matrix XGX' is symmetric.
Key concepts: Symmetric matrix, Square matrix, Mathematics, Centrosymmetric matrix, Matrix (chemical analysis), Pascal matrix, Nonnegative matrix, Inverse