Directory: stats/

.pro files

a_correlate_fix.pro

Perform the Cochrane-Orcutt Procedure to correct a time series data set for autocorrelated errors Applied Liner Statistical Models Neter Kutnre et al pp 509 use 1.65 at the upper limit to show no auto-correlation 0.01 confidence, 5 vars, and lots of data points y = b0 + b1[0]*x0 + b1[1]*x1 + etc

a_correlate_test.pro

Execute the Durbin-Watson Test for autocorrelation, this is a wrapper for dw_critical that makes it easy to do.

blom_position_test.pro

compute the blom position test.

bluenoise.pro

return a blue (f) noise spectra

bootstrap_mean.pro

compute the bootstrap mean

bootstrap_median.pro

compute the bootstrap median

boxcox.pro

confidence_band.pro

compute and optionally plot the confidence bands on a regression plot

dw_ac_test.pro

Perform the Durbin-Watson test for autocorrelation

dw_critical.pro

return the critical values for the Durbin-Watson test statistic from http://www.tau.ac.il/~yonar/econometrics/durbinwatson.pdf

ess_param_f_test.pro

compute the Fstar statistic that tells you if you can drop a term in a multiple regression this is from applied linear statistical models pp 268 fstar statistic

estimated_variance.pro

calculates the estimated variance in a set of residuals

fourplot.pro

create a fourplot for Exploratory Data Analysis http://www.itl.nist.gov/div898/handbook/eda/section3/4plot.htm The 4-plot is a collection of 4 specific EDA graphical techniques whose purpose is to test the assumptions that underlie most measurement processes.

geo_mean.pro

compute the geometric mean of an array

grubbs.pro

use the Grubbs' test for outlers on an input data set, see http://www.itl.nist.gov/div898/handbook/eda/section3/eda35h.htm

journal_20090324121025.pro

levene.pro

mode.pro

compute the moe of an array (most common element)

mse.pro

return the mean squared error of the inout residuls

normal_prob_plot.pro

make a normal probability plot

optbins.pro

parameter_t_test.pro

decide if a coefficient returned from a multiple regession is statistically different from zero this is from applied linear statistical models pp 232 (and other pages)

pinknoise.pro

return a pink (1/f) noise spectra

rednoise.pro

return a red noise spectra

resample.pro

create a uniform resampling of indicies from 0 to n_elements()-1

sixplot.pro

Produce a 6plot as defined in the NIST engineering statistics handbook http://www.itl.nist.gov/div898/handbook/eda/section3/6plot.htm The 6-plot is a collection of 6 specific graphical techniques whose purpose is to assess the validity of a Y versus X fit.

sse.pro

return the sum-squared error of the residuals

ssr.pro

calculate the sum squared residual.

stderr.pro

Compute the standard error of an array

violetnoise.pro

return a violet (f^2) noise spectra