MCS 142 Examination 3: Final Examination
- What: Examination focusing on statistical inference:
mainly chapters 7-9 and topics covered in class
- When: 10:30-12:30, Monday, May 26
- Where: Olin 318
Exam 3, the final examination,
will be a closed-book examination, but you will be allowed to use
one page of handwritten notes, your previous note cards,
the handout you received on hypothesis testing and confidence intervals,
your calculator, and tables provided.
Topics that exam problems may cover:
- Margin of error
- The margin of error is associated with a level of confidence.
See text pages 419, 425, 495, 543, 573, 589.
- Compare: standard error (= estimated standard deviation of
a statistic),
pp. 492-493, 573, 589, 591-592.
- Confidence intervals and tests of hypotheses for one population mean
- Known population standard deviation:
One-sample z statistic
- Unknown populatin standard deviation:
One-sample t statistic
- t distribution, degrees of freedom
- Matched-pairs t procedures
- Robustness of t procedures
- A nonparametric/distribution-free test for the median:
the sign test
The number of positive differences has a B(n, 1/2) distribution.
- Confidence intervals and tests of hypotheses for two population means
- Known population standard deviations:
Two-sample z statistic
- Unknown population standard deviations:
Two-sample t statistic
- Unknown, but known to be equal, population standard deviations:
Pooled two-sample t statistic
- Degrees of freedom of the two-sample t statistic
- Confidence intervals and tests of hypotheses for a single proportion
- Classical estimate of proportion = #successes/#observations.
- Wilson estimate = (#successes + 2)/(#observations + 4).
- Confidence intervals
- Large-sample z statistic for hypothesis tests
- Sample size for a desired margin of error
- Comparing two proportions
- Classical estimates and Wilson estimates
- Confidence intervals
- z statistic for
H0: p1 = p2
with pooled standard error
- Inference for two-way tables
- Displaying count data in two-way tables
- The chi-square statistic and the chi-square distribution
- Testing for association or dependence of row and column
variables in two-way tables
Previously covered topics that have a good chance of still being relevant:
- Numerical summaries
- Graphical summaries
- Normal tables, normal approximations
- The general procedure of hypothesis testing
and notions connected with tests of significance
- Confidence intervals and their interpretation
Suggested study and preparation:
- Review the summaries at end of each section of the text.
Make sure that you know and understand the terminology of statistical
inference.
- Look over homework problems and problems worked in class.
- Try some of the odd-numbered problems in chapters 7-9;
their answers are given in the back of our text.
- You may use either the Wilson estimates or the classical estimates for
proportions. The handout and your calculator use the classical estimates.