Comparison of the ability of the integral of the periodogram (NEV_i), the modified NEV equation (NEV_eq), and the bexvar NIV estimate (NEV_b), to accurately estimate the NIV of eROSITA-like light curves. This figure plots the absolute value of the difference between the measured value, NEV_m (which refers to either NEV_i, NEV_eq, or NEV_b), and the NIV (NIV), normalised by the NIV, as a function of the NIV. The lower the value of this parameter, the more accurate the estimate of the NIV is. Each data point represents one instance of a simulated light curve. The same light curves were used for computing NEV_i, NEV_eq, and NEV_b. Each panel represents a different count rate. The dependence on the number of bins is showcased via different symbols. Figure (a), (b), (c): Accuracy of the NIV estimates for white-, pink-, and red-noise light curves, respectively. Solid lines are plotted on top of the distribution of points, corresponding to the geometric mean of the distribution of log (|NEV_m – NIV|/NIV) as a function of log(NIV). Transparent areas indicate the part of the distribution corresponding to one standard deviation from the mean, in either direction.