Two estimators πΛ and πΛ are consistent and asymptotically normal.
Let π1,β¦,ππβΌπ.π.π.π€ππ(π) , for some π>0 .
Let πΛ=1ππ and πΜ =βln(ππ) , where ππ=1{ππ>1},π=1,β¦,π .
Find, their asymptotic variances π(πΛ) and π(πΛ)
To find the asymptotic variances of the two estimators πΛ and πΜ, we need to use their asymptotic properties.
1. Asymptotic Variance of πΛ:
The estimator πΛ = 1/Xn is consistent and asymptotically normal. Since π1,β¦,ππ βΌ π.π.π.π€ππ(π), we know that the sample mean πβ―β―β―n = (1/n) * Ξ£ππ is a consistent and asymptotically normal estimator of π.
The asymptotic variance of πΛ is given by:
π(πΛ) = [1 / (π^2)] * π(πβ―β―β―π)
Since π1,β¦,ππ βΌ π.π.π.π€ππ(π), the variance of πβ―β―β―π is given by:
π(πβ―β―β―π) = Var(1/n * Ξ£ππ)
= (1/n^2) * Ξ£Var(ππ) (due to independence)
= (1/n^2) * π * Var(π1)
= (1/n) * Var(π1)
= π(π1)
Therefore,
π(πΛ) = [1 / (π^2)] * π(π1)
2. Asymptotic Variance of πΜ:
The estimator πΜ = -ln(Yn), where ππ = 1{ππ > 1}.
Notice that ππ = 1 if ππ > 1 and ππ = 0 if ππ β€ 1.
This implies that Ξ£ππ is the count of ππ values greater than 1, which can be represented as n - Ξ£1{ππβ€1}.
Now, let's find the asymptotic variance of πΜ using the delta method:
We have:
πΜ = -ln(Yn)
Yn = Ξ ππ (since ππ = 1{ππ > 1})
= Ξ (1 - 1{ππ β€ 1})
Taking the natural logarithm on both sides:
ln(Yn) = ln(Ξ (1 - 1{ππ β€ 1}))
= Ξ£ln(1 - 1{ππ β€ 1})
Now, let's use the delta method to find the asymptotic variance:
π(πΜ) = [g'(π)]^2 * π(ππ(ππ))
where g(π) = -ln(π) and ππ(ππ) = Ξ£ln(1 - 1{ππ β€ 1}).
Differentiating g(π) with respect to π, we get:
g'(π) = -1 / π
The variance of ππ(ππ) can be approximated using the sample variance π^2:
π(ππ(ππ)) β (1/n) * π^2
Therefore,
π(πΜ) = [(-1/π)]^2 * (1/n) * π^2
Note: The above expressions for π(πΛ) and π(πΜ) are approximations based on the asymptotic properties of the estimators. In practice, it is recommended to use resampling methods or simulation to obtain more accurate estimates of the variances.
To find the asymptotic variances of the estimators Ξ»Μ and Ξ»Μ, we need to use some properties of consistent and asymptotically normal estimators.
First, let's find the asymptotic variance of Ξ»Μ:
The estimator Ξ»Μ = 1/Xβ is consistent and asymptotically normal.
To find the asymptotic variance of Ξ»Μ, we can use the Delta Method. The Delta Method states that if g(ΞΈ) is a differentiable function of the parameter ΞΈ, and βπ(ΞΈΜβΞΈ) converges in distribution to a normal distribution, then βπ(g(ΞΈΜ)βg(ΞΈ)) converges in distribution to a normal distribution with mean 0 and variance [g'(ΞΈ)]^2 times the variance of βπ(ΞΈΜβΞΈ).
In this case, g(ΞΈ) = 1/ΞΈ, and ΞΈ = Ξ». So, g'(ΞΈ) = -1/ΞΈΒ².
Using the Delta Method, the asymptotic variance of Ξ»Μ, denoted as V(Ξ»Μ), is given by:
V(Ξ»Μ) = [g'(Ξ»)]^2 * V(X)/n, where V(X) is the true variance of the random variable X.
In this case, X follows the exponential distribution with parameter λ, so V(X) = 1/λ².
Therefore, V(Ξ»Μ) = [(-1/λ²)]^2 * (1/λ²)/n = (1/Ξ»β΄)/n.
Now, let's find the asymptotic variance of Ξ»Μ:
The estimator Ξ»Μ = -ln(Yβ), where Yα΅’ = 1{Xα΅’ > 1}, i = 1,...,n.
To find the asymptotic variance of Ξ»Μ, we can use the Delta Method again. In this case, g(ΞΈ) = -ln(ΞΈ), and ΞΈ = P(X > 1).
Since Yα΅’ = 1{Xα΅’ > 1}, Yα΅’ follows a Bernoulli distribution with parameter P(X > 1). Therefore, the true variance of Yα΅’ is given by P(X > 1) * (1 - P(X > 1)).
Using the Delta Method, the asymptotic variance of Ξ»Μ, denoted as V(Ξ»Μ), is given by:
V(Ξ»Μ) = [g'(ΞΈ)]^2 * V(Y)/n, where V(Y) is the true variance of the random variable Yα΅’.
In this case, g'(ΞΈ) = -1/ΞΈ.
Therefore, V(Ξ»Μ) = [(-1/ΞΈ)]^2 * (ΞΈ(1 - ΞΈ))/n = (ΞΈ(1 - ΞΈ))/n.
However, we need to express the variance in terms of Ξ», not ΞΈ.
Since ΞΈ = P(X > 1) = 1 - P(X β€ 1), we have ΞΈ = 1 - e^(-Ξ»).
Substituting ΞΈ in the expression for V(Ξ»Μ), we get:
V(Ξ»Μ) = [(1 - e^(-Ξ»))(e^(-Ξ»))/n = [e^(-Ξ») - e^(-2Ξ»)]/n.
So, the asymptotic variances of Ξ»Μ and Ξ»Μ are:
V(Ξ»Μ) = (1/Ξ»β΄)/n
V(Ξ»Μ) = [e^(-Ξ») - e^(-2Ξ»)]/n