Let X0 be uniformly distributed on [0,1] and define the "tent map process" by Xn+1=1-2Xn-1, n[greater-or-equal, slanted]0. Let Mn=max(X1,...,Xn). We obtain the following results: For any integers n and k[greater-or-equal, slanted]1 we havewith the convention Cpq=0 if p<q and where [x] denotes the integer part of x (Theorem 1). For any [lambda]>0 we have limk--[infinity]...</q>

We approximate the distribution of scan statistics generated by some Bernoulli and Gaussian stationary 1-dependent models. We compare the results with the corresponding results provided by i.i.d. and Markov models.