Tools › Statistics › Descriptive and Outliers

Descriptive and Outliers

1 tool in the Statistics area.

Data Summary and Outliers

Descriptive statistics, normality tests and outlier tests for the columns of a pasted table: n, missing cells, mean, standard deviation, standard error and the 95 % t interval, median, quartiles (type 7 and the Tukey hinges), IQR, range, CV, MAD, 10 % trimmed mean, skewness and excess kurtosis with their standard errors; normality by Shapiro–Wilk, Anderson–Darling and the D'Agostino–Pearson K² omnibus; outliers by Tukey's fences, Grubbs' test (iterated), Dixon's ratio with Rorabacher's table and Rosner's generalized ESD test for several outliers; Pearson and Spearman correlations when two or more columns are chosen; a histogram with the normal curve, a box plot and a normal Q–Q plot. The default data are the NIST/SEMATECH e-Handbook heat flow meter calibration set (195 values, case study 1.4.2.8), whose printed mean, standard deviation, confidence interval, autocorrelation and Grubbs statistic the page reproduces, beside an illustrative small column with one planted outlier.

6 worked examples reproduced, listed on the tool page.

Other Statistics sections

Hypothesis Tests · Regression and Modelling · Distributions · Forecasting · Surveys and Sampling · Uncertainty and Monte Carlo · Probability