What is SMATR?
SMATR is a freely-available program used for fitting bivariate lines to data and for making inferences about such lines. A line can be fitted using standardised major axis (SMA), major axis (MA) or ordinary least squares regression (OLS) techniques. The routines available in SMATR allow you to:
- fit lines to groups of data using a SMA, MA or OLS fit
- construct exact confidence intervals about slope and intercept and compare these parameters to hypothesised values
- test for a common slope or elevation among lines fitted to several groups
- test for a shift along a common axis among lines fitted to several groups
- fit lines forced through the origin, appropriate for phylogenetic analyses
- incorporate measurement error into parameter estimates
For background, details about the statistics, and validation of these techniques see
Warton, DI, IJ Wright, DS Falster, and M Westoby (2006) Bivariate line-fitting methods for allometry. Biological Reviews 81(2):259-291 [link]