Ryan Boyden edited subsection_Intensity_Statistics_We_show__.tex  over 8 years ago

Commit id: 1ca0e244d9fc44d3619a9f8e0502d6343724cd38

deletions | additions      

       

\subsection{Intensity Statistics}  We show the colorplots for all intensity statistics in Figure ???. The PDF, Skewness, Kurtosis, PCA, and SCF produce similar colorplots, exhibiting the strongest sensitivities towards stellar mass-loss rates. The Kurtosis and Skewness statistics are yield  the most sensitive largest distances when a strong wind model is paired with a purely turbulent one, and a second largest distance when a strong wind model is compared to a weak wind model. For the PDF, PCA, and SCF, this distinction is not as clear, due to the changes in magnetic field strengths. PDF, Kurtosis, and Skewness also show smaller sensitivities towards changes in magnetic field strength under pairings of the same wind model, whereas PCA and SCF hardly show any (REWORD).  For these statistics, we find The Cramer Statistic, defined as a distance metric, exhibits behavior different from  that of  the largest distances are obtained by pairing other intensity statistics. We define the Cramer Statistic as the difference between two points in  a 2D array compared to the differences between two points within the array (cite Koch?) (check to see if the differences are weighted at all.). Looking at the Cramer Statistic colorplot, we find large distances when  strong wind model (W1) models are paired  witha  purely turbulent one (t0). The second largest values occur when comparing a strong wing model to a weak one (W2). The Kurtosis ones,  and Skewness plots clearly show this orde. Differences in magnetic field strengths appear to blur this distinction in our PDF, PCA, and SCF plots. relatively (significantly?) smaller values for all other pairs.   Argue which statistics are most useful?  Furthermore,   --Moment stats also slightly sensitive to T; PCA and SCF no!!!