In the 1960s Atkinson introduced an abstract algebraic setting for multiparameter eigenvalue problems. He showed that a nonsingular multiparameter eigenvalue problem is equivalent to the associated system of generalized eigenvalue problems. Many theoretical results and numerical methods for nonsingular multiparameter eigenvalue problems are based on this relation. We extend the above relation to singular two-parameter eigenvalue problems and show that the simple finite regular eigenvalues of a two-parameter eigenvalue problem and the common regular eigenvalues of the coupled generalized eigenvalue problem agree. Using the theory on the pencils of matrix polynomials we furthermore generalize the theory to the nonregular singular two-parameter eigenvalue problems. This enables one to solve a singular two-parameter eigenvalue problem by computing the common regular eigenvalues of the associated system of two singular generalized eigenvalue problems. There are various numerical methods for twoparameter eigenvalue problems, but all of them can only be applied to nonsingular problems. We develop a numerical method that can be applied to the singular two-parameter eigenvalue problems. It is based on the staircase algorithm for the extraction of the common regular part of two singular matrix pencils. We introduce the quadratic two-parameter eigenvalue problem (QMEP) and show that we can linearize it as a regular singular two-parameter eigenvalue problem. We present several transformations that can be used to solve the QMEP, by formulating an associated linear multiparameter eigenvalue problem. We also generalize the linearization to the polynomial twoparameter eigenvalue problem(PMEP). As an alternative approach to the linearization, we propose the transformation of the QMEP into a nonsingular five-parameter eigenvalue problem. We also consider several special cases of the QMEP, where some matrix coefficients are zero, which allows us to solve such QMEP more efficiently. We propose a Jacobi-Davidson type method for regular singular problems. We modify the Jacobi-Davidson type method for nonsingular two-parameter eigenvalue problem so that it can be applied to regular singular problems. The obtained algorithm can then be used to solve the problem obtained by linearizing the PMEP. If the dimension of matrices is large, then we cannot use the approach with linearization. If order of polynomials is small enough, then we can apply a Jacobi-Davidson type method directly to the polynomial system. This method is a generalization of the method for polynomial eigenvalue problems. We give some numerical results that illustrate the convergence of the introduced Jacobi-Davidson type methods.