On Ricci curvature of submanifolds in statistical manifolds of constant (quasi-constant) curvature (Q2129972): Difference between revisions

A Riemannian manifold \((\widehat B,\widehat g)\) with an affine connection \(\widehat\nabla\) is said to be a statistical manifold \((\widehat B,\widehat g,\widehat\nabla)\) if \(\widehat\nabla\) is a torsion free connection on \(\widehat B\) and the covariant derivative \(\widehat\nabla\widehat g\) is symmetric. A statistical manifold is a Riemannian manifold \((\widehat B,\widehat g)\) endowed with a pair of torsion-free affine connections \(\widehat\nabla\) and \(\widehat\nabla^*\) satisfying \[ (\widehat\nabla_{X_1}g)(Y_1,X_2)= X_1(\widehat g(Y_1,X_2))- \widehat g(\widehat\nabla_{X_1}Y_1,X_2)- \widehat g(Y_1,\widehat\nabla_{X_1}^* X_2)= 0, \] \[ X_1,Y_1\in\Gamma(T\widehat B),\quad X_2\in\Gamma(T^*\widehat B), \] where \(\Gamma(j)\) denote the space of sections of \((j)\) and \(\widehat\nabla^*\) is the conjugate connection. In this paper the authors study Ricci curvature of submanifolds in statistical manifolds of constant curvature. The proof is based on the analysis of a suitable constrained extremum problem. Using this method, the authors prove some results. Finally, the paper also contains several important examples. The exposition of the results is clear.