Abstract Scout
A univariate and multivariate data analysis tool with several classical and robust procedures (e.g., outlier testing, interactive 2D/3D graphics), making it a useful package for environmental and ecological applications. Scout can transform data, assess variable normality, produce histograms and Q-Q plots of raw data and principal component scores (PCSs), and produce scatter plots of raw data, PCSs, and discriminant scores. Additionally, Scout can be used to identify univariate or multivariate outliers; produce Q-Q plots of generalized distances; perform principal component, linear, and quadratic discriminant analyses; and compute and plot various statistical intervals, including the confidence interval for mean, prediction interval, and simultaneous confidence interval.