“…Choi and Shin [2] have suggested a change-point test for panel data of the form (X i,t ) i=1,..,N,t=1,...,T , and allow possible temporal correlation (that means Cov(X i,t , X i,s ) = 0 for s = t) or cross sectional correlation (Cov(X i,t , X j,u ) = 0 for i = j). They combine a test statistic introduced by Horváth and Hušková [3] once with circular bootstrap introduced by Politis and Romano [4] and once with the stationary bootstrap [5].…”