- Last Time:
- Review of measures of spread
- Especially the standard deviation (square root of the
variance)
- It has the right units to be the measuring stick
of spread - variance doesn't
- There are rules for using this measuring stick:
- Empirical rule (for bell-shaped
distributions): 68/95/99
- Chebysheff's Theorem: the (conservative)
0/75/89 rule
- Started chapter 8, section 3: the normal distribution
- The ugly probability density function (p. 231)
- both mean and standard deviation are visible from
the curve
- The mean and std dev determine the curve
completely.
- The empirical rule - the 68/95/99 rule - comes from the
normal curve
- Today:
- No homework due today
- Return homeworks: Chapter 2
- I need to see histograms! You can't talk about an
invisible histogram and expect me to understand: for example, each histogram
has its own bin size, which is important for the interpretation.
- Use descriptive titles on histograms (and on your
work in general). I frequently don't in class, because
I'm usually trying to make a statistical point with them, rather than producing
any "finished products". But what you turn in should be a "finished product!"
- Once again: If we give specific instructions on your
assignment schedule, then you are to follow those, rather than the instructinos
in the book!
- Questions about the homework associated with
chapter 4?
- Chapter 8: The normal distribution and probabilities
- Probabilities and the area under the curve
- What are probabilities?
- A calculator for normal probabilities
- The Z-score (or standard normal) transformation (p. 233)
- mean: 0; standard deviation: 1
- Hand out Z-Table
- Remember: symmetry!
- Calculating a Z-score (standardizing a normal curve: remember, all
normals are really the same - just shifted and stretched!)
- E.g., Joe Sixpack example
- Using the Z-score backwards
- Some examples
- Next time:
- Chapter 4 problems are due
- Start chapter 9 (sampling distributions)
Website maintained by Andy Long.
Comments appreciated.