Professional Documents
Culture Documents
Chapters 5-7
Q1
For a random variable T that is T (30) distributed, 1.3 1.3 = 0.80. Using this
result, derive an 80% confidence interval for the population mean given a random sample
with 31, 40 and /√ 10.
Q2
Suppose a 95% confidence interval for is (20, 30). Do you expect a wider, narrower,
or similar confidence interval in the following situations?
a) A 99% confidence interval is constructed.
b) The sample size is much larger.
c) The sample mean is much larger.
d) The standard deviation is much larger.
Q3
Suppose a random sample of 25 economists forecast economic growth for the next year. The
range of forecasts is from growth of -2.0 percent to growth of 3.5 percent with average 1.2
percent and with standard deviation of 2.0 percent. Anything with
an equality
a) Give a 95 percent confidence interval for the population mean forecast. sign is the null
b) Test at significance level 0.05 the claim that growth will be zero next year. State the
null and alternative hypotheses and your conclusion.
c) Test at significance level 0.05 the claim that the next year will be a recessionary year,
i.e. that growth will be negative. State the null and alternative hypotheses and your
conclusion.
We have following STATA Output – Use the relevant info forQ3-Q4
. display invttail(89, 0.1)
1.2911362
Command “inv” displays t values
. display invttail(89, 0.005)
2.6322042
Q4
The summary statistics for usual hours worked per week (variable hours) for a simple random
sample of women aged 30 years are the following
Variable Obs Mean Std. Dev Min Max
hours 90 32.81 19.73 0 90
a) Give a 95% confidence interval for population mean usual hours worked.
b) The claim is made that the population mean usual hours worked is 35 hours. Test this
claim at significance level 0.01. State the null and alternative hypotheses and your
conclusion.
c) The claim is made that the population mean usual hours worked is less than 35 hours.
Test this claim at significance level 0.10. State the null and alternative hypotheses and
your conclusion.
Q5
STATA gives the following output:
One-sample t test
------------------------------------------------------------------------------
Variable | Obs Mean Std. Err. Std. Dev. [95% Conf. Interval]
---------+--------------------------------------------------------------------
x | 90 197.9761 1.089684 10.33765 195.8109 200.1412
------------------------------------------------------------------------------
mean = mean(x) t = 1.8134
Ho: mean = 196 degrees of freedom = 89
Ha: mean < 196 Ha: mean != 196 Ha: mean > 196
Pr(T < t) = 0.9634 Pr(|T| > |t|) = 0.0731 Pr(T > t) = 0.0366
(a) Suppose that we want to test : versus : . Show that this null
hypothesis cannot be rejected, and explain in plain English what that means.
(b) Suppose that we want to test : versus : . Show that this null
hypothesis can be rejected, and explain in plain English what that means.
(c) So we reject that , while we do not reject that . That seems contradictory.
What is going on here?
Q7
Assume the usual setup, with a random sample of size n taken from a normal distribution
with mean and variance . However, this time we choose not to use the sample mean
but rather the sample median M as an estimator for .