Description Usage Arguments Details Value References See Also
Wrapper function sdt
Creates a 'Signal Detection Theory' vector
Creates a 'Signal Detection Theory' vector
1 2 3 4 5 6 7 |
hi |
numeric; hits / true positives |
... |
further parameter |
fa |
numeric; false alarms / false positives |
mi |
numeric; misses / false negatives |
cr |
numeric; correct rejection / true negatives |
criterion |
logical vector |
prediction |
logical vector |
This function returns: hitrate (sensitivity/TPR), specifity (true negative rate/SPC), false alarm rate (fall-out/FPR), false discovery rate (FDR), an estimated d' (qnorm(hitrate)-qnorm(false alarm rate)) and the MCC, the "Matthews correlation efficient", c-bias (c < 0 -> liberal; c > 0 -> conservative).
Some results are adjusted, to make them calculatable. If one of the contingency-values hi
, fa
, mi
or cr
equals zero,
all of them will gain .25: Sdt(1, 0, 2, 4)
equals Sdt(1.25, .25, 2.25, 4.25)
.
The denominator of the Matthews correlation coefficient is adjusted to 1 if (hi + fa) == 0
, (hi + mi) == 0
, (fa + cr) == 0
or (cr + mi) == 0
.
numeric vector with signal-detection values
numeric vector with signal-detection values
http://kangleelab.com/signal detection theory.html http://en.wikipedia.org/wiki/Matthews_correlation_coefficient
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.