Stat 322 – HW 1

Due Friday, Jan. 12

 

1) Complete the Stat 322 Questionnaire in Blackboard (under the Assignments button).

 

2) Find 2 uses of statistics (e.g., in the news, on the web, from other courses).  For each, write a paragraph summarizing the use of statistics (what did they do) and then asking a few questions about it, e.g., what additional information do you want to know?  Be sure to cite the sources and dates of the articles.

 

3) (p. 127) problem 54. If you use Minitab, include a well-labeled copy of your output.

 

Binomial with n = 10 and p = 0.01

x  P( X <= x )

2     0.999886

Binomial with n = 10 and p = 0.05

x  P( X <= x )

2     0.988496

Binomial with n = 10 and p = 0.1

x  P( X <= x )

2     0.929809

Binomial with n = 10 and p = 0.2

x  P( X <= x )

2     0.677800

Binomial with n = 10 and p = 0.25

x  P( X <= x )

2     0.525593

 

 

 

Binomial with n = 10 and p = 0.01

x  P( X <= x )

1     0.995734

Binomial with n = 10 and p = 0.05

x  P( X <= x )

1     0.913862

Binomial with n = 10 and p = 0.1

x  P( X <= x )

1     0.736099

Binomial with n = 10 and p = 0.2

x  P( X <= x )

1     0.375810

Binomial with n = 10 and p = 0.25

x  P( X <= x )

1     0.244025

 

 

(d)

Binomial with n = 15 and p = 0.01

x  P( X <= x )

2     0.999584

Binomial with n = 15 and p = 0.05

x  P( X <= x )

2     0.963800

Binomial with n = 15 and p = 0.1

x  P( X <= x )

2     0.815939

Binomial with n = 15 and p = 0.2

x  P( X <= x )

2     0.398023

Binomial with n = 15 and p = 0.25

x  P( X <= x )

2    0.236088

 

 

(e)

 

n=15 curve does the best job of being high when p is low and low when p is high.  The P(X  1, n = 10) has high probabilities throughout, but we don’t want to accept the batch when p is large.

 

4) (p. 172) problem 34.

.6827

 

.9987

Second one wins hands down!

 

4) (p. 242) problem 52, make sure you justify your steps. Hint: You might want to consider calculating the probability in terms of the average lifetime for the package instead.

We want to find t such that P(T > t) < .05 where the random variable T represents the total lifetime of all batteries.  We can instead find  such that P( > ) < .05 and convert.

Since the lifetimes follow a normal distribution with m = 10 and s = 1, then  follows a normal distribution with mean 10 and standard deviation equal to s/sqrt(4) = .5.

 

To find the 95th percentile, this corresponds to z = 1.645.  So our observation needs to be 1.645 standard deviations above the mean.  So we would need to observe a sample mean of 10 + 1.645(.5) = 10.8 hours.  This corresponds to a total lifetime of 43.2 hours.

 

5) (p. 298) problem 20 using formula 7.11 (p. 295) and/or Minitab (including your output in the latter case)

 

n =4722           = .15           z = 2.576 for 99% confidence

 

Our sample size is quite large so we can use the “wald procedure”:

 ±  z  = .15 ±  2.576sqrt(.15(.85)/4722) = .15 ±  .0134 = (.1366, .1634)

 

Minitab:

 

We are 99% confident that the proportion of all American youngsters who are seriously overweight is between .137 and .163.