Skip to main content

Advertisement

Table 3 Results of fitting a parametric survival regression using the logistic distribution to the unique URLs.

From: A cross disciplinary study of link decay and the effectiveness of mitigation techniques

Variable Value p 5% 95%
(Intercept) 5.22 3.3E-30 4.46 5.97
Log2(URL published) 3.57 1.4E-17 2.88 4.25
depth -1.46 7.0E-32 -1.66 -1.25
Log2(TimesCited + 1) 0.25 2.8E-04 0.13 0.36
Funding text present 3.43 2.8E-11 2.59 4.28
Domain
au 4.53 1.5E-04 2.56 6.49
be 3.31 1.9E-02 0.99 5.64
ca 4.88 1.7E-06 3.20 6.56
ch 6.45 7.2E-08 4.48 8.42
cn 1.50 1.3E-01 -0.13 3.13
com 6.02 2.2E-18 4.89 7.16
de 5.74 6.1E-16 4.57 6.91
dk 7.66 5.7E-07 5.14 10.18
edu 3.77 1.6E-13 2.93 4.61
es 3.05 5.4E-03 1.25 4.85
fr 3.65 6.6E-07 2.44 4.85
gov 5.51 1.2E-15 4.38 6.64
il 5.92 3.6E-04 3.19 8.65
in 4.78 2.2E-04 2.65 6.91
it 5.51 1.4E-08 3.91 7.11
jp 5.07 8.0E-09 3.62 6.51
kr -3.35 2.0E-02 -5.73 -0.97
net 7.01 4.2E-11 5.26 8.76
nl 6.78 1.1E-06 4.49 9.07
org 8.10 2.4E-36 7.04 9.16
ru 3.90 2.3E-03 1.80 6.01
se 1.71 2.4E-01 -0.69 4.12
tw 1.64 1.7E-01 -0.33 3.61
uk 4.49 4.2E-12 3.42 5.56
Source
Bioinformatics -2.04 5.7E-03 -3.25 -0.83
BMC Bioinformatics 2.69 3.9E-05 1.62 3.77
BMC Genomics 0.88 4.7E-01 -1.13 2.89
Comp. Physics Comm. -4.00 3.0E-05 -5.57 -2.42
Genome Research 0.56 7.1E-01 -1.92 3.04
Nucleic Acids Research 1.28 8.6E-04 0.65 1.91
PLoS ONE -0.39 8.0E-01 -2.95 2.18
Zoological Studies 16.42 2.2E-15 13.01 19.83
  1. Positive numbers indicate longer median lifetimes. Much like a logistic model, coefficients can be added to the intercept value (after multiplying in the case of numeric predictors) to obtain a median lifetime. For example, the median expected lifetime for a URL published once, with depth 0, whose publishing article had 1 citation, no funding text, domain au and published in a Journal not listed (ie- in the default) would be: (Intercept) 5.22 + Log2(1)*3.57 + 0*-1.46 + Log2(1+1)*0.25 + 0*3.43 + 4.53 = 10 years