Problem Set 2 - Estimation and Hypothesis Testing

Problem Set 2 - Estimation and Hypothesis Testing#

(Check Canvas for due dates and office hours)

  1. Answer case-by-case (use common notations).

    • If the population follows \(\mathcal{B}(N,p)\).

      • What is the distribution of \(\bar{x}\)?

      • What is \(\hat{p}\)?

      • If a sample of size 50 from a binomial distribution of 5 trials is known to have \(\bar{x} = 3.6\),

        • What is \(\hat{p}\)?

        • Provide 90% lower confidence bound on \(\hat{p}\).

    • If the population follows \(Poisson(\lambda)\).

      • What is the distribution of \(\bar{x}\)?

      • What is \(\hat{\lambda}\)?

  2. Assuming the poulation follows a normal distribution \(\mathcal{N}(\mu,\sigma^2)\),

    • What are the distributions of

      • \(\bar{x}\)

      • \(\frac{\bar{x} - \mu}{\sigma/\sqrt{n}}\)

      • \(\frac{\bar{x} - \mu}{s/\sqrt{n}}\)

      • \(\frac{(n-1)s^2}{\sigma^2}\)

    • What are estimates of

      • \(\hat{\mu}\)

      • \(\hat{\sigma^2}\)

      • \(\hat{\sigma}\)

    • If we choose a sample of size 9 from a normal distribution, provide confidence bounds on \(\hat{\mu}\) for each case.

      • 95% two-sided bounds with \(\bar{x} = 1.5\) and \(\sigma^2 = 4\).

      • 95% two-sided bounds with \(\bar{x} = 1.5\) and \(s^2 = 4\).

      • 95% upper confidence bound with \(\bar{x} = 1.5\) and \(\sigma^2 = 4\).

      • 95% lower confidence bound with \(\bar{x} = 1.5\) and \(s^2 = 4\).

  3. If we choose a sample of size 9 from a normal distribution with \(\mu = 1.5\) and \(\sigma^2 = 4\).

    • What is the probability (as percentage) that \(\bar{x} \geq 2.327\)?

    • What is the probability (as percentage) that \(\bar{x} \leq 0.26\)?

    • What is the probability (as percentage) that \( -0.737 \leq \bar{x} \leq 3.737\)?

  4. If we choose a sample of size 9 with \(s^2 = 4\) from a normal distribution with \(\mu = 1.5\) and unknown \(\sigma^2\).

    • What is the probability (as percentage) that \(\bar{x} \geq 2.327\)?

    • What is the probability (as percentage) that \(\bar{x} \leq 0.26\)?

    • What is the probability (as percentage) that \( -0.737 \leq \bar{x} \leq 3.737\)?

    • Comment on the difference between answers of this question and the previous.

  5. Answer Qualitatively: If we choose a sample of size 9 with \(\bar{x} = 1.5\), arrange the three statements in order from more likely to less likely. (NO NEED TO CALCULATE ANY VALUES).

    • (1) Sample belongs to a population with \(\mu = 2\) and \(\sigma^2 = 0.01\)?

    • (2) Sample belongs to a population with \(\mu = 2\) and \(\sigma^2 = 1\)?

    • (3) Sample belongs to a population with \(\mu = 0\) and \(\sigma^2 = 1\)?

  6. Forming Hypothesis (just write \(H_0\) and \(H_a\), no need to analyze): You have a sample of 9 softdrink cans with mean width \(\bar{x} = 1.5\) (inches) from a normal distributed manufacturing process (with huge number of cans) with unknown \(\mu\). Convert the following plain language statements into null (\(H_0\)) and alternative hypotheses (\(H_a\)).

    • Can the manufacturing process have a mean width that is different from \(\hat{\mu} \,\, (=\bar{x})\)?

    • Can the manufacturing process have a mean width that is greater than \(3.737\)?

    • Can the manufacturing process have a mean width that is less than \(0.26\)?

  7. If we choose a sample of size 9 with \(\bar{x} = 1.5\) from a normal distribution with unknown \(\mu\). Consider Type-I error rate (\(\alpha\)) as 5% or 10%.

    • What is the p-value for \(H_0: \mu = 2\); \(H_a: \mu \neq 2\) with known \(\sigma^2 = 0.01\)? What will be the result of hypothesis testing?

    • What is the p-value for \(H_0: \mu = 2\); \(H_a: \mu \neq 2\) with \(s^2 = 0.01\) and unknown \(\sigma^2\)? What will be the result of hypothesis testing?

    • What is the p-value for \(H_0: \mu = 2\); \(H_a: \mu < 2\) with known \(\sigma^2 = 1\)? What will be the result of hypothesis testing?

    • What is the p-value for \(H_0: \mu = 2\); \(H_a: \mu < 2\) with \(s^2 = 1\) and unknown \(\sigma^2\)? What will be the result of hypothesis testing?

    • What is the p-value for \(H_0: \mu = 0\); \(H_a: \mu < 0\) with known \(\sigma^2 = 1\)? What will be the result of hypothesis testing?

    • What is the p-value for \(H_0: \mu = 0\); \(H_a: \mu < 0\) with \(s^2 = 1\) and unknown \(\sigma^2\)? What will be the result of hypothesis testing?

  8. Textbook Problem 4.3: The service life of a battery used in a cardiac pacemaker is assumed to be normally distributed. A random sample of 10 batteries is subjected to an accelerated life test by running them continuously at an elevated temperature until failure, and the following lifetimes (in hours) are obtained: 25.5, 26.1, 26.8, 23.2, 24.2, 28.4, 25.0, 27.8, 27.3, and 25.7.

    • The manufacturer wants to be certain that the mean battery life exceeds 25 hours. What conclusions can be drawn from these data (use \(\alpha\) = 0.05)?

    • Construct a 90% two-sided confidence interval on mean life in the accelerated test.

  9. Textbook Problem 4.14: A random sample of 200 printed circuit boards contains 18 defective or nonconforming units. Estimate the process fraction nonconforming.

    • Test the hypothesis that the true fraction nonconforming in this process is 0.10. Use \(\alpha= 0.05\). Find the P-value.

    • Construct a 90% two-sided confidence interval on the true fraction nonconforming in the production process.

  10. Textbook Problem 4.9: The output voltage of a power supply is assumed to be normally distributed. Sixteen observations taken at random on voltage are as follows: 10.35, 9.30, 10.00, 9.96, 11.65, 12.00, 11.25, 9.58, 11.54, 9.95, 10.28, 8.37, 10.44, 9.25, 9.38, and 10.85.

    • Test the hypothesis that \(\sigma^2 = 1.0\) using \(\alpha= 0.05\).

    • Construct a 95% two-sided confidence interval on \(\sigma^2\).

    • Construct a 95% upper confidence interval on \(\sigma\).