deletions | additions       diff --git a/section_Stastical_Comparisons_label_comparisons__.tex b/section_Stastical_Comparisons_label_comparisons__.tex  index c9f6ebb..3cdbf82 100644  --- a/section_Stastical_Comparisons_label_comparisons__.tex  +++ b/section_Stastical_Comparisons_label_comparisons__.tex 
...
For both runs, the most probable range of kurtosis values are close to zero. However, run W1T2t0.2 has higher probabilities at positive values, and is 15 percent less likely to have a kurtosis of zero than that of run W2T2t0. The Skewness PDFs indicate a similar trend, in which run W1T2T0.2 is 10 percent less likely to have a skewness of zero. On small scales, the integrated intensity distributions of our two runs behave similar to a Gaussian, though run W1T2t0.2 has a greater tendency to diverge from this behavior than run W2T2t0.2 does.  \subsections{Principal \subsection{Principal  Component Analysis} (The Principal Component Analysis (PCA) is one of the most sensitive statistics in our fiducial comparison.) We present our runs' velocity channel covariance matrices in Figure ???. After launching winds, the covariances for velocity channels between -2 and 2 km/s (the channels of strongest intensity???) increase in a particular behavior: features at the center of run W2T2t0's covariance matrix are augmented and further separated (as shown in run W1T2t0.2' covariance matrix). Covariances outlining the features also increase.