Calculate mahalanobis distance (null and alternative MSEs) and statistical power for all MPs in an MSE

Usage

PRBcalc(
  MSE_null,
  MSE_alt,
  tsd = c("Cat", "Cat", "Cat", "Ind", "ML"),
  stat = c("slp", "AAV", "mu", "slp", "slp"),
  dnam = c("C_S", "C_V", "C_M", "I_S", "ML_S"),
  res = 6,
  alpha = 0.05,
  plotCC = FALSE,
  removedat = FALSE,
  removethresh = 0.025
)

Arguments

MSE_null: An object of class MSE representing the null hypothesis
MSE_alt: An object of class MSE representing the alternative hypothesis
tsd: Character string of data types: Cat = catch, Ind = relative abundance index, ML = mean length in catches
stat: Character string defining the quantity to be calculated for each data type, slp = slope(log(x)), AAV = average annual variability, mu = mean(log(x))
dnam: Character string of names for the quantities calculated
res: Integer, the resolution (time blocking) for the calculation of PPD
alpha: Probability of incorrectly rejecting the null operating model when it is valid
plotCC: Logical, should the PPD cross correlations be plotted?
removedat: Logical, should data not contributing to the mahalanobis distance be removed?
removethresh: Positive fraction: the cumulative percentage of removed data (removedat=TRUE) that contribute to the mahalanobis distance

Value

A list object with two hierarchies of indexing, first by MP, second has two positions as described in Probs: (1) mahalanobis distance, (2) a matrix of type 1 error (first row) and statistical power (second row), by time block.

References

Carruthers, T.R, and Hordyk, A.R. In press. Using management strategy evaluation to establish indicators of changing fisheries. Canadian Journal of Fisheries and Aquatic Science.

Author

T. Carruthers