1: \begin{abstract} Extreme value statistics (EVS) concerns the study of the statistics of the maximum or the minimum 

2: of a set of random variables. This is an important problem for any time-series and has applications in climate, 

3: finance, sports, all the way to physics of disordered systems where one is interested in the statistics of the 

4: ground state energy. While the EVS of `uncorrelated' variables are well understood, little is known for strongly 

5: correlated random variables. Only recently this subject has gained much importance both in statistical physics and 

6: in probability theory. In this review, we will first recall the classical EVS for uncorrelated variables and discuss the

7: three universality classes of extreme value limiting distribution, known as the Gumbel, Fr\'echet and

8: Weibull distribution. We then show that,

9: for weakly correlated random variables with a finite correlation length/time, the limiting extreme value distribution

10: can still be inferred from that of the uncorrelated variables using a renormalisation group-like argument. Finally, we consider 

11: the most interesting examples of strongly correlated variables for which there are very few exact results for the EVS. We discuss

12: few examples of such strongly correlated systems (such as the Brownian motion and the eigenvalues of

13: a random matrix) where some analytical progress can be made. We also discuss other observables related to extremes, such as the density of near-extreme events, time at which an extreme value occurs, order and record statistics, etc. 

14: \end{abstract}

15: