Certifiable randomness from temporal correlations

Leggett-Garg inequalities (LGI) are constrains on certain combinations of temporal correlations obtained by measuring one and the same system at two different instants of time. The usual derivations of LGI assume \emph {macroscopic realism per se } and \emph {noninvasive measurability}. We derive these inequalities under a different set of assumptions, namely the assumptions of \emph{predictability} and \emph{no signaling in time}. As a novel implication of this derivation, we show that LGI can be used to certify randomness in a device independent way.