Limiting spectral distribution of high-dimensional noncentral Fisher matrices and its analysis

Fisher matrix is one of the most important statistics in multivariate statistical analysis. Its eigenvalues are of primary importance for many applications, such as testing the equality of mean vectors, testing the equality of covariance matrices and signal detection problems. In this paper, we establish the limiting spectral distribution of high-dimensional noncentral Fisher matrices and investigate its analytic behavior. In particular, we show the determination criterion for the support of the limiting spectral distribution of the noncentral Fisher matrices, which is the base of investigating the high-dimensional problems concerned with noncentral Fisher matrices.