URL: http://en.diveintodeeplearning.org/chapter_crashcourse/sampling.html

# Sampling

**mli**#1

In section 2.6.3, how are the following plots derived?

plt.semilogx(y,(variance ** 0.5) * np.power(y,-0.5) + mean,‘r’)

plt.semilogx(y,-(variance ** 0.5) * np.power(y,-0.5) + mean,‘r’)

I couldn’t understand why are we plotting with y-axis as (variance**0.5) * np.power(y,-0.5) + mean

Edit: My understanding is that we are plotting the graph to see if the sample is falling in the range of [mean - SE , mean + SE] with respect to sample size. Is this correct?

where SE = Standard Error = sigma / sqrt(n)