Follow:
|
Here are some tips as you prepare for the Final Exam
|
- First, as has frequently been the case these days, OMIT the section on Multiple Linear Regression in Lesson 11 of my book, and all of Lesson 12 (The Sign Test). That means you should only study up to the end of #3 in Lesson 11 (Inferences for
Regression).
- In addition, a prof has announced that they have cut out the hypothesis for correlation this term. So that means the only hypothesis tests from Lesson 11 you should be prepared for are Hypothesis for Zero Slope using t with df= n-2 and Hypothesis for Zero Slope using Anova F-Test with df= 1, n-2.
- Make sure you are familiar with all four confidence interval formulas from Lesson 11.
- Confidence Interval
for the Intercept
- Confidence Interval for the slope
- Confidence interval for a prediction at a given x
- Confidence interval for the average y at a given x
- Do the long answer section FIRST on the final exam. Make sure you have looked at all the practice long answer questions on the practice exams (note that I have a solutions and audio offer for these below). Topics to be especially ready for as long answer
(these are just my opinion; I have no inside information):
- A Type I Error, Type II Error, Power question where you need to compute the x-bar decision rule and make an alpha/beta table as I teach in Lesson 3 of my book.
- A Probability question where you will use either a Two-Way Table or a Venn Diagram to set it up and solve questions, including a conditional probability question, as I teach
in Lesson 6 of my book. There was also a good example in Assignment 3 (links below).
- A Chi-Square Problem. Probably a Goodness-of-Fit question, probably using the Binomial Distribution (you are given the binomial table in the booklet of tables on the exam).
- A Inferences for Linear Regression problem where they give you the least-squares regression line, and give you a partially-filled in Anova
table, and they give you the sum of (xi - x-bar)^2 which is extremely helpful for computing all the standard errors used in the confidence intervals. You then need to fill in the Anova Table and know how it can tell you r-squared (SSM/SST) and Se, the standard deviation of the residuals (MSE is the variance of the residuals). They will then ask you to compute some confidence intervals and test hypotheses about the zero slope.
- Never stop to think on a
Multiple-Choice Question. If you don't understand it, guess, and move on. If you do understand it, but think it will take too long, guess and move on. Come back to the tough ones only after you have completed the exam. The most important thing is that you gave yourself a chance to answer all the questions you do understand.
- Never do a multiple-choice question TWICE! If you know you are wrong because your answer
is not one of the choices, too bad! Guess, pick the closest, and move on. Never waste time trying to figure out what went wrong. You have other questions to do! Come back to it with fresh eyes after you have completed the exam.
- Best wishes and STUDY HARD!
The department posted SOLUTIONS for Assignment 1 (these are not my solutions). Click here.The department posted SOLUTIONS for Assignment 2 (these are not
my solutions). Click here.The department posted SOLUTIONS for Assignment 3 (these are not my solutions). Click here. |
|
