Pursuit of unknown Periods in chronobiology.

Most series of chronobiological measurements show more than one deflection. The question is which deflections are the main ones. The surest and swiftest method for solving this problem is the classical Fourier analysis on the basis of the least squares method. In the case of equidistant prescribed measuring points, an orthogonal system of linear equations is obtained from which the Fourier coefficients of all sine and cosine coefficients are derived, so that the Fourier spectrum can be plotted by computer or drawn by hand. In case of non-equidistant points, the system of linear equations is non-orthogonal but uniquely soluble. If there is more than one point assigned to each given value of argument (i.e., time), the problem of grouped Fourier regression must be solved. Mathematically, this is the same as in the case of non-equidistant data. By means of statistical tests the complete system of Fourier coefficients can be reduced so that only significant periods remain.