Scribd
Upload
6 views2 pages

Unbiased Estimators in Statistical Methods

The document contains practice problems related to statistical estimation methods, focusing on unbiased estimators for various distributions including exponential, uniform, normal, and beta distributions. It also explores the Neyman Fisher Factorization Theorem for finding sufficient statistics. The problems require demonstrating the unbiasedness of certain estimators and finding sufficient statistics based on given random samples.

Uploaded by

shruti.prabhakar2002
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
6 views2 pages

Unbiased Estimators in Statistical Methods

The document contains practice problems related to statistical estimation methods, focusing on unbiased estimators for various distributions including exponential, uniform, normal, and beta distributions. It also explores the Neyman Fisher Factorization Theorem for finding sufficient statistics. The problems require demonstrating the unbiasedness of certain estimators and finding sufficient statistics based on given random samples.

Uploaded by

shruti.prabhakar2002
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

MTL390: Statistical Methods

January 27, 2025

Practice Problems 7
Theory of Estimation
1. Let X1 , X2 , . . . Xn be a random sample from an exponential distribution with p.d.f.
 
1 x
fX (x) = exp − ;x > 0
β β
Pn
Show that X̄ = i=1 Xi /n is an unbiased estimator of β.
n+1
2. Let X1 , X2 , . . . Xn be a random sample from U (0, θ); θ > 0. Show that n X(n) and 2X̄ are both
unbiased estimators of θ.
3. Let X1 , X2 , . . . Xn be a random sample from an exponential distribution with p.d.f.

f (x) = β exp(−βx); x > 0


Show that X̄ is an unbiased estimator of 1/β.
 Pn 2
Let X1 , X2 , . . . Xn be a random sample from N θ, θ2 , θ > 0. Show that ( i=1 Xi ) /n(n + 1) and
4. P
n 2 2
i=1 Xi /2n are both unbiased estimators of θ .

5. Let X1 , X2 , . . . Xn be a random sample from P (θ); θ > 0. Find an unbiased estimator of θe−2θ .
6. Let X1 , X√ . Xn be a random sample from B(1, θ); 0 ≤ θ ≤ 1. (a) Show that the estimator
2, . . P
1
n+ n X
T (X ) = 2 √i=1 i is not unbiased θ ?
∼ n+ n

(b) Show that limn→∞ E(T X ) = θ.
∼
(An estimator satisfying the condition in (b) is said to be unbiased in the limit)

7. X1 , . . . , Xn be a random sample from N µ, σ 2 , µ ∈ R, σ ∈ R+ . Find unbiased estimators of µ/σ 2
and µ/σ.
8. Let X1 , X2 , . . . Xn be a random sample from B(1, θ); 0 ≤ θ ≤ 1. Find an unbiased estimator of
θ2 (1 − θ).
9. Using Neyman Fisher Factorization Theorem, find a sufficient based on a random sample X1 , X2 , . . . Xn
from each of(the following distributions
1
exp − αx

α if x > 0
(a) fα (x) =
0 otherwise.
(
exp(−(x − β)) if x > β
(b) fβ (x) =
0 otherwise.
(  
1 (x−β)
exp − α if x > β
(c) fα,β (x) = α
0 otherwise.
(  
1 (log xi −µ)2
√ exp − 2σ 2 if x > 0
(d) fµ,σ (x) = xσ 2π
0 otherwise.
(
1
−θ/2 ≤ x ≤ θ/2
(e) fθ (x) = θ
0 otherwise

1
10. Let X1 and X2 be independent random samples with densities f1 (x1 ) = θe−θx1 and f2 (x2 ) =
2θe−2θx2 as the respective p.d.f.s where θ > 0 is an unknown parameter and 0 < x1 , x2 < ∞.
Using Neyman Fisher Factorization Theorem find a sufficient statistic for θ.

11. Let X1 , . . . , Xn be a random sample with densities


(
exp(iθ − x) if x ≥ iθ
fXi (x) =
0 otherwise .

Using Neyman Fisher Factorization Theorem find a sufficient statistic for θ.


12. Let X1 , X2 , . . . Xn be a random sample from a Beta(α, β) distribution (α > 0, β > 0)
ShowQnthat
(a) Qi=1 Xi is sufficient for α if β is known to be a given constant.
n
(b) Qi=1 (1 −QXi ) is sufficient for β if α is known to be a given constant.
n n
(c) ( i=1 Xi , i=1 (1 − Xi )) is jointly sufficient for (α, β) if both the parameters are unknown.
13. Let T and T ∗ be two statistic such that T = ψ (T ∗ ). Show that if T is sufficient then T ∗ is also
sufficient.
14. X1 , . . . , Xn be a random sample from U (θ − 1/2, θ + 1/2), θ ∈ R. Find a sufficient statistic for θ.

15. Let X1 , . . . , Xn be independent random variables with Xi (i = 1, 2, . . . , n) having the probability


density function
(
iθe−iθxi xi > 0
fi (xi ) =
0 otherwise

Find a sufficient statistic for θ.

You might also like