(a) Solution:
During second catch and release, 10 trout were tagged out of 100 trout that were caught. Assuming that for every 100 trout caught 10% will always have tags, total trout in the lake approximately = \(10 * 100 = 1000\).
(b) Solution:
Using hypergeometric distribution, the following is the probability function of \(X\).
\(P(X=x) = h(N,k,n,x) = \frac{{k \choose x}{{N-k} \choose {n-x}}}{N \choose n}\)
where \(N\) is population estimate
\(k\) is sample size of trout caught first time\((k = 100)\)
\(n\) is sample size of trout caught second time\((n = 100)\)
\(x\) is tagged trout in caught second time\((x = 10)\)
\(P(N) = h(N,100,100,10) = \frac{{100 \choose 10}{{N-100} \choose {100-10}}}{N \choose 100}\) = \(\frac{{100 \choose 10}{{N-100} \choose 90}}{N \choose 100}\).
(c) Solution:
Using growth quotient of hypergeometric model, _Maximum likelihood estimate(\(MLE\))
\(Q(N) = \frac{P(N)}{P(N-1)} > 1\)
where \(P(N) = \frac{{k \choose x}{{N-k} \choose {n-x}}}{N \choose n}\)
\(P(N - 1) = \frac{{k \choose x}{{N-1-k} \choose {n-x}}}{N-1 \choose n}\)
\(\frac{P(N)}{P(N-1)} > 1\)
\(P(N) > P(N-1)\)
\(= \frac{{k \choose x}{{N-k} \choose {n-x}}}{N \choose n} > \frac{{k \choose x}{{N-1-k} \choose {n-x}}}{N-1 \choose n}\)
Solving the equation will result in
\(= N(N - k -n + x) < (N - x)(N-n)\)
\(= N < \frac{kn}{x}\)
Substituting values for \(k, n\ and\ x\)
\(= N < \frac{100 * 100}{10}\) = \(1000\)
Let’s use R function to get Maximum likelihood estimate(\(MLE\))
library(knitr)
#initiate values
N1<- 100*100/10
k<- 100
n<- 100
x<- 10
PN<- matrix(NA, nrow=10, ncol=2)
#calculate upper values
for (i in 0:4){
N<- N1 + i
#hypergeometric distribution
output<- (choose(k,x) * choose(N-k, n-x))/choose(N,n)
PN[6+i,1]<- N
PN[6+i,2]<- output
}
#calculate lower values
for (i in 1:5){
N<- N1 - i
#hypergeometric distribution
output<- (choose(k,x) * choose(N-k, n-x))/choose(N,n)
PN[6-i,1]<- N
PN[6-i,2]<- output
}
kable(PN, format="pandoc", align="l", col.names = c("N", "P(N)"), caption = "Hypergeometric distribution")
| N | P(N) |
|---|---|
| 995 | 0.1389680 |
| 996 | 0.1389749 |
| 997 | 0.1389801 |
| 998 | 0.1389835 |
| 999 | 0.1389853 |
| 1000 | 0.1389853 |
| 1001 | 0.1389836 |
| 1002 | 0.1389801 |
| 1003 | 0.1389750 |
| 1004 | 0.1389682 |
At \(N = 1000\) and \(N = 999, P(N) = 0.1389853\), hence it is maximum value, for all other values of \(N, P(N)\) is lower.
Therefore, Maximum likelihood estimate(\(MLE\)) = \(1000\) Or \(999\).
References: