Permutation of Pointwise Linear Regression (PoPLR): calculation of the p-value

Description

For details see [1]

Usage

1
poplr_pstat( vf, porder, type = "slr", sl_test = NULL )

Arguments

vf

visual-field data. It can be td or pd as well

porder

order of permunations of visual-fields sensitivities. Each row contains a permutation of tests from vf

type

Type of regression statistic: slr for simple linear regression and rank for Spearman correlation coefficient. Default is slr

sl_test

values for the 1-tailed hypothesis test for each location. The reference values are not restricted, but they should be either zero (was there any progression?) or negative (was the progression greater than test value?). Default is NULL

Value

the function retunts two different structures depending on whehter the analysis is linear regression (type = "slr" or Spearman rank correlation type = "rank"). For slr analysis, the sturcture consists of four matrices with data: pval (p-value at each permutation and location of the significance of the 1-tailed hypothesis test specified by sl_test for each location), se (standard error), sl (slope), and int (intercept). For rank analysis, the sturcture consists of two matrices: pval (p-value at each permutation and location of the significance of the 1-tailed hypothesis test specified by sl_test for each location) and rho (the Spearman rank correlation)

Author(s)

Ivan Marin-Franch

References

[1] N. O'Leary, B. C. Chauhan, and P. H. Artes. Visual field progression in glaucoma: estimating the overall significance of deterioration with permutation analyses of pointwise linear regression (PoPLR). Investigative Ophthalmology and Visual Science, 53, 2012

See Also

poplr, hist_poplr, poplr_cstat, vflayout_poplr, vfplot_poplr

Examples

1
res <- poplr( vf91016right )

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.