On the other hand, when comparing time series that maintained their autocorrelation structures we smoothed the raw, monthly climate index time series with a 5-point finite impulse response filter until both series being compared had similar spectral energy above the annual harmonic.
The properties of the real and simulated series are compared from both fractal and chaotic perspective using a new multifractal deviation factor, and a crossing analysis based on correlation dimension and non linear prediction error, respectively.
The measured and simulated free surface time series are compared and analyzed using both one- and two-dimensional short-time Fourier transform.
Kenyon's suit, which also lays out similarities between characters, says that when the series are compared "in many respects, the elements are virtually identical".
Numerical results evaluated using a moderate number of terms in the series are compared against finite difference solutions of the problem and against previous results reported in the literature, showing excellent agreement in both cases.
The time series of As, V, Cr, and Fe in all springs are plotted in Fig. 5, together with the water pH values only from the four springs affected by deep influx (S1, S2, S3, and S4; Fig. 1a, Supplementary Table S3), and these series are compared with earthquake magnitudes in the same period.
Tomogram slices close to a Mimivirus capsid layer (corresponding to the boxed area in Fig. 2) built from specified numbers N of tilt series are compared in Fig. 3.
When the strength developments of mortars in three series are compared, as expected, the highest strengths were observed in the series of 100% OPC as binder and the lowest strengths were observed in the series with 30% cement replacement by fly ash.
Simulated and measured time series are compared in Figure 3.
In cross-correlation analysis, two time series are compared using the Pearson correlation coefficient with an increasing lag of one series with respect to the other.
