In this paper, we examine observational constraints on the power law cosmology; essentially dependent on two parameters $H_0$ (Hubble constant) and $q$ (deceleration parameter). We investigate the constraints on these parameters using the latest 28 points of H(z) data and 580 points of Union2.1 compilation data and, compare the results with the results of $\Lambda$CDM. We also forecast constraints using a simulated data set for the future JDEM, supernovae survey. Our studies give better insight into power law cosmology than the earlier done analysis by Kumar [arXiv:1109.6924] indicating it tuning well with Union2.1 compilation data but not with H(z) data. However, the constraints obtained on $<H_0>$ and $<q>$ i.e. $H_0$ average and $q$ average using the simulated data set for the future JDEM, supernovae survey are found to be inconsistent with the values obtained from the H(z) and Union2.1 compilation data. We also perform the statefinder analysis and find that the power-law cosmological models approach the standard $\Lambda$CDM model as $q\rightarrow -1$. Finally, we observe that although the power law cosmology explains several prominent features of evolution of the Universe, it fails in details.

A modified Mimetic gravity (MMG) is proposed as a generalization of general relativity. The model contain a physical metric which is function of an auxiliary (unphysical) metric and a Lyra's metric. We construct different kinds of conformally invariant models in different levels of the expansion parameter $\lambda$. This model phenomenologically has been extended to higher order forms. Cosmology of a certain class of such models has been investigated in details. A cosmological solution has been proposed in inhomogeneous form of scalar field. For homogenous case, energy conditions are widely investigated. We have shown that the system evaluated at intervals shorter than a certain time $T_c$ meets all the energy conditions.