The Gamma distribution Gamma(πΌ,π½) with paramters πΌ>0 , and π½>0 is defined by the density
ππΌ,π½(π₯)=π½πΌΞ(πΌ)π₯πΌβ1πβπ½π₯,for allπ₯β₯0.
The Ξ function is defined by
Ξ(π )=β«β0π₯π β1πβπ₯ππ₯.
As usual, the constant π½πΌΞ(πΌ) is a normalization constant that gives β«β0ππΌ,π½(π₯)ππ₯=1.
In this problem, let π1,β¦,ππ be i.i.d. Gamma variables with
π½=1πΌfor some πΌ>0.
That is, π1,β¦,ππβΌGamma(πΌ,1πΌ) random variables for some πΌ>0 . The pdf for ππ is therefore
ππΌ(π₯)=1Ξ(πΌ)πΌπΌπ₯πΌβ1πβπ₯/πΌ,for all π₯β₯0.
What is the limit, in probability, of the sample average πβ―β―β―β―β―π of the sample in terms of πΌ ?
πβ―β―β―β―β―πββββπββπ = ?
----------------------------------------------------------------------------------------------------
Use the result from the previous problem to give a consistent estimator πΌΜ of πΌ in terms of πβ―β―β―β―β―π .
(Enter barX_n for πβ―β―β―β―β―π )
πΌΜ =?
----------------------------------------------------------------------------------------------------
For the Delta method to apply, at what value of π₯ does π need to be continuously differentiable? (Your answer should be in terms of πΌ .)
π₯=?
----------------------------------------------------------------------------------------------------
What is its asymptotic variance of πΌΜ ?
π(πΌΜ )=
(a) alpha^2
(b) barX_n^(1/2)
(c1) alpha^2
(c2) Normal Distribution
(c3) ?
(d)?
I got green tick with alpha^2 for c1
Isolve=[ 1.89, 2.37]
Iplug-in=[ 1.88, 2.36]
Using the previous part, find confidence intervals for πΌ with asymptotic level 90% using both the βsolving" and the βplug-in" methods. Use π=25 , and πβ―β―β―β―β―π=4.5 .
(Enter your answers accurate to 2 decimal places. Use the Gaussian estimate π0.05β1.6448 for best results.)
Isolve=[ , ]
Iplug-in=[ , ]
c3 and d?
To find the limit in probability of the sample average πβ―β―β―β―β―π, we can use the Law of Large Numbers. The Law of Large Numbers states that the sample average of a sequence of independent and identically distributed random variables converges in probability to the expected value of the random variable.
In this case, since π1, π2, ..., ππ are i.i.d. Gamma random variables with parameters (πΌ, 1/πΌ), the expected value of each ππ is πΌ/π½ = πΌ. Therefore, the limit in probability of πβ―β―β―β―β―π as π approaches infinity is π = πΌ.
To obtain a consistent estimator πΌΜ of πΌ in terms of πβ―β―β―β―β―π, we can simply use the sample average itself. Therefore, πΌΜ = πβ―β―β―β―β―π.
For the Delta method to apply, the function π needs to be continuously differentiable at a certain value π₯. In this case, we need to find the value of π₯ at which π is continuously differentiable. However, the function π is not provided in the given question, so we cannot determine the specific value of π₯.
Finally, to find the asymptotic variance of πΌΜ, denoted as π(πΌΜ), we would need more information about the estimator πΌΜ and the distribution of the random variables π1, π2, ..., ππ. However, this information is not provided in the question, so we cannot determine π(πΌΜ).