The Cholesky factorization can fail for nearly singular matrices in trying to take the square root of something that became negative due to rounding error. Can the Cholesky factorization from the Levinson-Durbin algorithm fail in the same manner or is it different? What is the point of failure?
Already registered? Login
Not Account? Sign up
Enter your email address to reset your password
Back to Login? Click here