State Markov's inequality and Chebyshev's inequality.
Let G(2)(n,p) denote the probability space of bipartite graphs with vertex classes U={1,2,…,n} and V={−1,−2,…,−n}, with each possible edge uv(u∈U,, v∈V) present, independently, with probability p. Let X be the number of subgraphs of G∈G(2)(n,p) that are isomorphic to the complete bipartite graph K2,2. Write down EX and Var(X). Hence show that p=1/n is a threshold for G∈G(2)(n,p) to contain K2,2, in the sense that if np→∞ then a. e. G∈G(2)(n,p) contains a K2,2, whereas if np→0 then a. e. G∈G(2)(n,p) does not contain a K2,2.
By modifying a random G∈G(2)(n,p) for suitably chosen p, show that, for each n, there exists a bipartite graph H with n vertices in each class such that K2,2⊂H but e(H)⩾43(3n−1n)2.