PROPHET StatGuide: Expanded List of Topics


Jump to a section on this page: Descriptive statistics, testing distributions, comparison tests, nonparametric comparison tests, comparing regression lines, frequency and contingency tables, variance tests, survival analysis, exploratory data analysis, fitting models to data.
Descriptive statistics:

Descriptive statistics (sample size, number missing, mean, median, standard deviation, etc.)


Testing distributions:

Testing distributional goodness of fit (normality tests, Kolmogorov-Smirnov test, chi-square goodness of fit test)


Comparison tests:

Testing equality of means or location (t test, analysis of variance [ANOVA], sign test [one-sample and two-sample paired], signed rank test [one-sample and two-sample paired], rank sum test, Kruskal-Wallis test, Friedman's test, Compare Samples, Cochran's Q test, McNemar's Q test)


Nonparametric comparison tests:

Nonparametric tests of location or spread (sign test [one-sample and two-sample paired], signed rank test [one-sample and two-sample paired], rank sum test, Kruskal-Wallis test, Friedman's test, Compare Samples, Cochran's Q test, McNemar's Q test, Ansari-Bradley test)


Comparing regression lines:

Comparing simple linear (straight-line) regression lines (analysis of covariance [ANCOVA])


Frequency and contingency tables:

Frequency and contingency tables (categorical analysis) (creating frequency tables, cross-tabulation [creating contingency tables, measures of association], contingency table analysis [chi-square test for independence, Fisher's exact test])


Variance tests:

Testing equality of variances or dispersions (chi-square test for variance, F test, Levene's test, Bartlett's test, Ansari-Bradley test)


Survival analysis:

Survival analysis (analyzing and comparing survival functions) (life tables, Kaplan-Meier plot, comparing survival curves [Mantel-Cox test, Gehan-Breslow test, Tarone-Ware test])


Exploratory data analysis:

Exploratory data analysis (EDA) (discovering patterns in data) (coded tables, median polish, letter values, repeated-median regression)


Fitting models to data:

Model fitting (simple linear (fitting a line: b0 + b1*x),
multiple linear (fitting a linear combination of x terms: b0 + b1*x1 + b2*x2 + ... + bn*xn),
polynomial (fitting a polynomial, b0 + b1*x + b2*x**2 + ... + bn*x**n),
non-linear (fitting an arbitrary function f(x), e.g., (vmax*x) / (km+x),
dose*exp(-k*t)),
logistic (fitting the logit of y by a linear combination of x terms:
b0 + b1*x1 + b2*x2 + ... + bn*xn))


Back to the top of this page.

Examine the glossary.

Do a keyword search of PROPHET StatGuide.

Send us your feedback.

Back to StatGuide home page.

Last modified: March 11, 1997

©1996 BBN Corporation All rights reserved.