Scribd
Upload
2 views2 pages

STAT 2202 Tutorial 1

The document discusses properties of good estimators in statistical analysis, focusing on unbiasedness and efficiency of various estimators for population mean and variance. It includes examples of estimators for the mean of juice packet weights, variance from a sample, and a parameter from a specific distribution, along with calculations for bias, variance, and mean squared error. Additionally, it addresses sufficient statistics for a given distribution, providing a comprehensive overview of estimator evaluation.

Uploaded by

yugonid
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
2 views2 pages

STAT 2202 Tutorial 1

The document discusses properties of good estimators in statistical analysis, focusing on unbiasedness and efficiency of various estimators for population mean and variance. It includes examples of estimators for the mean of juice packet weights, variance from a sample, and a parameter from a specific distribution, along with calculations for bias, variance, and mean squared error. Additionally, it addresses sufficient statistics for a given distribution, providing a comprehensive overview of estimator evaluation.

Uploaded by

yugonid
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

TUTORIAL-PROPERTIES OF GOOD ESTIMATORS

1. A company manufactures fruit juice packets. Suppose the weight of juice


packets follows a normal distribution with mean weight 𝜇 grams and standard
deviation 𝜎 grams. To estimate the average weight of the fruit juice packets, the
quality control inspector randomly selects three packets and records their
weights 𝑋1 , 𝑋2 , 𝑎𝑛𝑑 𝑋3 . Based on these observations, the inspector proposes
two estimators for the population mean 𝜇:
𝑋1 +𝑋2 +𝑋3 𝑋1 +𝑋2 𝑋3
𝑇1 = and 𝑇2 = + .
3 4 2
a. Are both estimators unbiased for 𝜇?
b. Which one of them is more efficient?

2. Let 𝑋1 , 𝑋2 , 𝑋3 , … , 𝑋𝑛 be a random sample from a population with mean 𝜇 and


1
variance 𝜎 2 . The sample mean is 𝑋̅ = ∑𝑛𝑖=1 𝑋𝑖 . The estimator
𝑛
1
𝜎̂ 2 = 𝑛 ∑𝑛𝑖=1(𝑋𝑖 − 𝑋̅)2 is sometimes used to estimate 𝜎 2 .

a. Find the bias of 𝜎̂ 2 .


1
b. Hence show that the estimator 𝑆 2 = 𝑛−1 ∑𝑛𝑖=1(𝑋𝑖 − 𝑋̅)2 is an unbiased
estimator of 𝜎 2 .

3. Let 𝑋1 , 𝑋2 , 𝑋3 , … , 𝑋𝑛 be a random sample from a distribution with the


probability density function:
1
𝑓𝑋 (𝑥) = { 𝜃 (𝑥 − ) + 1, 𝑓𝑜𝑟 0 ≤ 𝑥 ≤ 1
2
0 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
where 𝜃 ∈ [−2, 2] is an unknown parameter.
̂ 𝑛 = 12𝑋̅ − 6 , where 𝑋̅ is the sample mean.
Suppose the estimator of 𝜃 is Θ
a. Find 𝐸[𝑋].
b. Find 𝑉𝑎𝑟(𝑋).
̂ 𝑛 = 12𝑋̅ − 6 is an unbiased estimator of 𝜃.
c. Show that Θ
̂ 𝑛 is a consistent estimator of 𝜃.
d. Determine whether Θ
̂𝑛 .
e. Find the mean squared error of Θ
4. Let 𝑋1 , 𝑋2 , 𝑋3 , … , 𝑋𝑛 be a random sample from a distribution with mean
𝐸[𝑋𝑖 ] = 𝜃 and Var[𝑋𝑖 ] = 𝜎 2 . Consider the following two estimators for 𝜃:
̂ 1 = 𝑋1 and Θ
Θ ̂ 2 = 𝑋1 +𝑋2+ 𝑋3+ …+𝑋𝑛.
𝑛
a. Find 𝑀𝑆𝐸( Θ
̂ 1 ) and 𝑀𝑆𝐸( Θ
̂ 2 ).
b. Hence, show that for 𝑛 > 1, 𝑀𝑆𝐸( Θ
̂ 1 ) > 𝑀𝑆𝐸( Θ
̂ 2 ).

5. Let 𝑌1 , 𝑌2 , … , 𝑌𝑛 be independent and identically distributed observations from a


−𝑦2
𝑦
distribution with density 𝑓(𝑦; 𝜃) = 𝜃 𝑒 2𝜃 , 𝑦 ≥ 0, 𝜃 > 0. Find a sufficient statistic
for 𝜃.

You might also like