Scribd
Upload
18 views2 pages

Probability and Statistics Problem Set VI

Uploaded by

Shriram A ep23b019
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
18 views2 pages

Probability and Statistics Problem Set VI

Uploaded by

Shriram A ep23b019
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

MA2040: Probability, Statistics and Stochastic Processes

Problem Set-VI
Sivaram Ambikasaran
April 3, 2019

1. Let X be the height of a randomly chosen individual from a population. In order to estimate the mean
and variance of X, we observe a random sample X1 , X2 , . . . , X7 . Thus, Xi ’s are i.i.d. and have the same
distribution as X. We obtain the following values (in centimeters): 166.8, 171.4, 169.1, 178.5, 168.0, 157.9, 170.1
Find the values of the sample mean, the sample variance, and the sample standard deviation for the
observed sample.
2. Prove the following:
• If Θ̂ is an unbiased estimator for Θ, then so is Θ̂ + W , where W is a zero mean random variable.
 
• If Θ̂1 is an estimator for Θ such that E Θ̂1 = aΘ + b, where a 6= 0, show that

Θ1 − b
Θ̂2 =
a
is an unbiased estimator for Θ.
3. Let {Xi }ni=1 be a random sample from a uniform distribution (0, Θ), where Θ is unknown. Define the
estimator
Θ̂ = max{X1 , X2 , . . . , Xn }
• Find the bias
of Θ̂  
2
• Find the E Θ̂ − Θ

• Is Θ̂ a consistent estimator of Θ?
4. Let X1 , X2 , . . . , Xn be a random sample from a geometric distribution with parameter p, where p is
unknown. Find the maximum likelihood estimator of p based on the random sample {Xi }ni=1 .
5. Let X1 , X2 , . . . , Xn be a random sample from a uniform distribution (0, Θ), where Θ is unknown. Find
the maximum likelihood estimator of Θ based on the random sample {Xi }ni=1 .
2
6. Let X1 , X2 , X3 , X4 be a random sample from a normal distribution with mean µ and σ = 3. If s
2
denotes the sample variance, and P s ≤ k = 0.05, find k.
 
7. If X ∼ N µ = 10, σ 2 = 25 and X1 , X2 , . . . , X501 is a random sample from X, find E s2 .
8. If T ∼ t19 , find C so that P (|T | ≤ X) = 0.95.
9. Let X1 , X2 , . . . , X11 be a random sample of size 11 from a normal distribution with unknown mean µ
11
X √ √
and variance σ 2 = 9.9. If Xi = 132, find k so that [12 − k 0.9, 12 + k 0.9] is a 90% confidence
i=1
interval for µ.

1
10. If X1 , X2 , . . . , Xn is a random sample from a normal population with variance σ 2 = 25, how large must
the sample size be so that the length of a 95% confidence interval for µ is 1.96?
11. A building contractor usually gets his light bulbs from shop A, where he knows that the bulbs last
for an expected time of 100 months with a standard deviation of 3 months. A new shop B contacts
the building contractor and say that he can supply bulbs which will last more than those from shop
A maintaining the same standard deviation. The building contractor has to now decide whether he
needs to make an order of 1 lakh light bulbs from shop A or shop B. To do this, he asks 16 customers
who purchased from shop B and finds out that that these light bulbs lasted for an average of 103
months. The contractor doesn’t like to take risk and decides to use hypothesis testing with a very low
significance value of 1% to decide. Will he purchase from shop A or shop B? You may assume that
the times are normally distributed.

12. A dealer purchases an idli grinder from manufacturer A, where he knows that the expected time to
grind 2 kgs of flour to a desired consistency is 15 minutes with a standard deviation of 3 minutes. A
new manufacturer B contacts the dealer and proposes that his grinders can grind faster than those
from manufacturer A with the same standard deviation of 3 minutes. The dealer samples 10 grinders
from B and finds that they take 14 minutes to do the same process. The dealer is willing to take risk
and decides to use hypothesis testing with a reasonably high significance value of 10% to decide. Will
he purchase from shop A or shop B? You may assume that the times are normally distributed.
13. A premium gold ball production line must produce all of its balls of 45.75 grams in order to get the top
rating. Samples are drawn hourly and checked. If the production line gets out os sync with a statistical
significance of more than 1%, it must be shut down and repaired. A sample of 18 balls has a mean of
45 grams and a sample standard deviation of 2 grams. Should the production line be repaired?
14. In the first problem, the second shop owner (B) says that his bulbs last for an average of 105 months
with a standard deviation of 3 months. If the contractor wants his type I error to be 1% and type
II error to be 5%, how many samples should he check and what is his critical value for accepting and
rejecting the null hypothesis?

Common questions

Powered by AI

With the null hypothesis H0: μ = 45.75 grams and α = 0.01, determine whether the production line requires repair. Use a z-test given n = 18, μ = 45.75, x̄ = 45, and σ = 2. Calculate the z-score and check if it falls beyond the critical value for a two-tailed test. If it does, reject H0, indicating significant deviation requiring repair .

To ensure a confidence interval (CI) length of 1.96, use the formula for the CI: L = 2*Z*(σ/√n). Set L = 1.96, Z = 1.96 for 95% CI, and σ = 5. Solving for n gives: n = (2*5*1.96 / 1.96)^2 = 100. Thus, a sample size of 100 is needed .

The concept of linear unbiased estimators supports this. If ˆΘ is an unbiased estimator of Θ, adding any zero-mean random variable, W, to it (ˆΘ + W) maintains the unbiasedness because the expectation of W is zero: E(ˆΘ + W) = E(ˆΘ) + E(W) = Θ + 0 = Θ. Thus, ˆΘ + W retains the property of being unbiased .

To find the critical value, start by setting α = 0.01 for Type I error. Use the distribution of the test statistic to find the critical region that results in a rejection of the null hypothesis. For Type II error (β = 0.05), identify the probability of inaccurately retaining the null hypothesis when the alternative is true. Calculate β using the N-P test approach, find sample mean under the alternative, relate it to the critical region, and set the power (1-β). Use this to solve for the critical value alongside balancing with power .

The contractor sets up a null hypothesis H0: μA = 100 months against the alternative hypothesis H1: μB > μA. Using the sample mean of 103 from shop B and known population standard deviation of 3 months, perform a one-tailed test with α = 0.01. Calculate the z-score and compare it to the critical value from the standard normal distribution. If the z-score exceeds the critical value, reject H0, suggesting bulbs from shop B last significantly longer .

The estimator ˆΘ overestimates Θ because it takes the maximum of the sample values, which introduces positive bias. This bias arises because each observation underestimates the true maximum more than it overestimates it, causing max{X1, X2, ..., Xn} to systematically be lower than the true Θ as it fails to account for extreme value occurrences outside the sample range .

To determine the critical value k for a sample variance, use the chi-square distribution as s²(n-1)/σ² follows a χ² distribution with (n-1) degrees of freedom. From the requirement P(s² ≤ k) = 0.05, find k such that the cumulative probability up to this value under χ² with given degrees of freedom equals 0.05. Use tables or statistical software to find the corresponding k .

Calculating the MLE is important as it provides a method for estimating parameters in a statistical model that maximizes the likelihood of observing the given data. For a uniform distribution, the MLE of Θ is the largest observed data point, as this maximizes the likelihood function. It provides an efficient and unbiased estimate especially in larger samples due to the properties of consistency, asymptotic normality, and optimality .

The dealer establishes H0: μA = 15 minutes against H1: μB < μA. Calculating the sample mean (14 minutes) and using the known standard deviation, perform a one-tailed test. With a significance level of 0.10, determine the critical t-value. Compute the t-statistic using the sample data. If the calculated t-statistic falls in the critical region, reject H0, indicating grinders from manufacturer B are faster .

First, calculate the sample mean (X̄) by summing all data points and dividing by the number of data points: X̄ = (166.8 + 171.4 + 169.1 + 178.5 + 168.0 + 157.9 + 170.1) / 7 = 169.7 cm. Next, calculate the sample variance (s²) using the formula: s² = Σ(Xi - X̄)² / (n - 1), yielding approximately 49.99 cm². Finally, the sample standard deviation (s) is the square root of the sample variance, approximately 7.07 cm.

You might also like