The Euclid space telescope will measure the shapes and redshifts of galaxies to reconstruct the expansion history of the Universe and the growth of cosmic structures. Estimation of the expected performance of the experiment, in terms of predicted constraints on cosmological parameters, has so far relied on different methodologies and numerical implementations, developed for different observational probes and for their combination. In this paper we present validated forecasts, that combine both theoretical and observational expertise for different cosmological probes. This is presented to provide the community with reliable numerical codes and methods for Euclid cosmological forecasts. We describe in detail the methodology adopted for Fisher matrix forecasts, applied to galaxy clustering, weak lensing and their combination. We estimate the required accuracy for Euclid forecasts and outline a methodology for their development. We then compare and improve different numerical implementations, reaching uncertainties on the errors of cosmological parameters that are less than the required precision in all cases. Furthermore, we provide details on the validated implementations that can be used by the reader to validate their own codes if required. We present new cosmological forecasts for Euclid. We find that results depend on the specific cosmological model and remaining freedom in each setup, i.e. flat or non-flat spatial cosmologies, or different cuts at nonlinear scales. The validated numerical implementations can now be reliably used for any setup. We present results for an optimistic and a pessimistic choice of such settings. We demonstrate that the impact of cross-correlations is particularly relevant for models beyond a cosmological constant and may allow us to increase the dark energy Figure of Merit by at least a factor of three.