We address three genuine phantom dark energy models where each of them induces the particular future events known as Big Rip, Little Rip and Little Sibling of the Big Rip. The background models are fully determined by a given dark energy equation of state. We first observationally constrain the corresponding model parameters that characterise each paradigm using the available data of supernova type Ia, Cosmic Microwave Background and Baryonic Acoustic Oscillations by using a Markov Chain Monte Carlo method. The obtained fits are used to solve numerically the first order cosmological perturbations. We compute the evolution of the density contrast of (dark) matter and DE, from the radiation dominated era till a totally DE dominated universe. Then, the obtained results are compared with respect to $\Lambda$CDM. We obtain the predicted current matter power spectrum and the evolution of $f\sigma_8$ given by the models studied in this work. Finally, the models are tested by computing the reduced $\chi^2$ for the `Gold2017' $f\sigma_8$ dataset.

In order to study the effect of the anti de Sitter/ conformal field theory correspondence (AdS/CFT) on the primordial inflationary era, we consider a universe filled with a tachyon field in a slow-roll regime. In this context, the background and perturbative parameters characterising the inflationary era are related to the standard one by correction terms. We show a clear agreement between the theoretical prediction and the observational data for the above mentioned model. The main results of our work are illustrated for an exponential potential. We show that, for a suitable conformal anomaly coefficient, AdS/CFT correspondence might leave its imprints on the spectrum of the gravitational waves amplitude with a tensor to scalar ratio, $r$, of the perturbations compatible with Planck data.

We present a new cosmological event, which we named the little sibling of the big rip. This event is much smoother than the big rip singularity. When the little sibling of the big rip is reached, the Hubble rate and the scale factor blow up but the cosmic derivative of the Hubble rate does not. This abrupt event takes place at an infinite cosmic time where the scalar curvature explodes. We show that a doomsday \'a la little sibling of the big rip is compatible with an accelerating universe, indeed at present it would mimic perfectly a LCDM scenario. It turns out that eventhough the event seems to be harmless as it takes place in the infinite future, the bound structures in the universe would be unavoidably destroyed on a finite cosmic time from now. The model can be motivated by considering that the weak energy condition should not be abusibely violated in our Universe, and it could give us some hints about the status of recently formulated nonlinear energy conditions.

It is well known that the holographic Ricci dark energy can induce some future doomsdays in the evolution of the universe. Here we analyse the possible avoidance of those doomsdays by invoking a modification to general relativity on the form of curvature effects.

We present an holographic model of the Dvali-Gabadadze-Porrati scenario with a Gauss-Bonnet term in the bulk. We concentrate on the solution that generalises the normal Dvali-Gabadadze-Porrati branch. It is well known that this branch cannot describe the late-time acceleration of the universe even with the inclusion of a Gauss-Bonnet term. Here, we show that this branch in the presence of a Gauss-Bonnet curvature effect and an holographic dark energy with the Hubble scale as the infra-red cutoff can describe the late-time acceleration of the universe. It is worthwhile to stress that such an energy density component cannot do the same job on the normal Dvali-Gabadadze-Porrati branch (without Gauss-Bonnet modifications) nor in a standard 4-dimensional relativistic model. The acceleration on the brane is also presented as being induced through an effective dark energy which corresponds to a balance between the holographic one and geometrical effects encoded through the Hubble parameter.