Ratio estimators are in principle biased, although the bias may be fairly
small if the number of stations is large. More importantly, the longer
soaks would be weighted more in the ratio estimator, which I don't think is
what you want.
Dvora
On Thu, Jan 28, 2016 at 3:19 PM, Mark Terceiro (NOAA Federal) <
mark.terceiro@xxxxxxxx> wrote:
Most trawl surveys use a version the former - calculate catch per tow, the
calculate the variance of those 'catch rates.'
On 1/28/2016 3:15 PM, Lee, Laura wrote:
Hi all!
I’m having a discussion with one of our biologists about the best way to
calculate an index for one of our gill-net surveys. The survey typically
employs a 24-hour soak, though sometimes, largely due to weather, soaks may
last up to four days. This presents us with at least two options for
calculating the index. Option 1 would be to standardize the catches to
one-day soaks (divide the catch by the number of days fished) and take the
arithmetic average of those standardized catches. Option 2 would be to
calculate the index as a ratio estimator. That is, sum all the catches and
divide by the sum of days fished. The difference is subtle and trends are
almost identical, but the calculation of variance will depend on the option
selected.
Is one of these options more appropriate than the other? I’d appreciate
any thoughts on this. Thanks so much!
Cheers,
Laura
*Laura M. Lee*
*Senior Stock Assessment Scientist*
Division of Marine Fisheries
Department of Environmental Quality
252 808 8094 office
252 726 6062 fax
Laura.Lee@xxxxxxxxxx
3441 Arendell Street
P.O. Box 769
Morehead City, NC 28557-0769
*Email correspondence to and from this address is subject to the*
*North Carolina Public Records Law and may be disclosed to third parties.*