さあ！諸君！捕鯨問題だ！

＞モデルは正規分布を前提にしているが、現実には０−１００％以外の確率密度分布は０。
＞まあ、信頼度が低い（常識の範囲以下の信頼限界値）っつうだけのことだから、
＞わざわざ分布モデル変えて統計処理するほどの意味はねえな。

モデルは正規分布を前提になんかしてないですよ。
調査捕鯨対象をランダムサンプリングするとして、現場レベルでの正規分布を
設定してますが、出てくるデータが正規分布になるなんてことは前提になってません。

以下、本当に統計学が解ってる人のための付録：
REP. INT. WHAL. COMMN 40. 1990 p．531 SC/41/Ｏ15
”Estimation of Natural Mortality Coefficient of Whales from the
Estimates of Abundance and Age Composition Data Obtained
from Research Catches”
Syoiti Tanaka
Tokyo University of Fisheries, 4-5-7, Konan, Minato-ku, Tokyo 108, Japan

ABSTRACT
Japanese research on the Antarctic minke whale is intended to provide
unbiased estimates of both population abundance and
population age composition. A method to estimate the age-dependent
natural mortality coefficient, ＜Ma＞, from time series of the above
data is presented. A preliminary analysis of the precision of the
estimates is given. ＜Ma＞ is divided into two components, the average
level for the whole population, and age dependency (deviation from
the average). Estimation of the former needs both time series
estimates of the abundance and age composition whereas the latter
needs only the age composition. The precision of the average level
for the whole population is almost wholly determined by the variability
in sighting surveys, and can be improved considerably by
increasing the number of repetitions or the number of years of the survey.
Estimation of the age dependency for each age individually
is difficult unless the sample size is large. However, ＜Ma＞ is expected
to change rather smoothly with age, and the fitting of a
curve on the relationship between ＜Ma＞ and age would present the
entire pattern of variability of ＜Ma＞ depending on age, and thus
provide the value of 　 age dependency for each age with a smaller
sample size. It is concluded that useful estimates can be obtained
with realistic sample size and repeated sampling. A brief discussion
on possible sources of bias and future studies needed is given.

原文では年齢依存の死亡（率）係数＜Ma＞ に括弧がついてないですけど、
活字の関係で読みづらいから＜＞つけました。