1: \begin{abstract}

2: We prove that  if a rectangular $r\times n$ matrix with uniformly

3: small entries and approximately orthogonal rows is applied to the

4: independent standardized random variables with uniformly bounded

5: third moments, then the empirical CDF of the resulting partial sums

6: converges to the  normal CDF with probability one. This implies

7: almost sure convergence of empirical periodograms, almost sure

8: convergence of spectra  of circulant and reverse circulant matrices,

9: and almost sure convergence of the CDF's generated from independent

10: random variables by independent random orthogonal matrices.

11: 

12: For special trigonometric matrices, the speed of the almost sure

13: convergence is described by the normal approximation and by the

14: large deviation principle.

15: \end{abstract}

16: