Difference between revisions of "Lectures: Stat 202 Soc Poly Fall 2016"

From Sean_Carver
Jump to: navigation, search
(September 12, 2016)
Line 45: Line 45:
 
* Homework: [[Media:Stat202_2015S_HW5.pdf|Homework 5]] was assigned today with a due date of September 21, 2016.
 
* Homework: [[Media:Stat202_2015S_HW5.pdf|Homework 5]] was assigned today with a due date of September 21, 2016.
 
* Next class: we will discuss density curves and start probability.
 
* Next class: we will discuss density curves and start probability.
 +
 +
== September 14, 2016 ==
 +
 +
* Today we reviewed standard deviation, transformations, density curves, the relationship between density curves and histograms.  We discussed the fact that all density curves describe a distribution, but to be a density curve: necessarily the area under the curve must be 1 and whole density curve must be on or above the vertical axis.  We talked about bell curves and the relationship between bell curves and the normal distribution (bell curves are the density curves for normal distributions).  Bell curves have a very specific shape that depends on only two parameters: mean and standard deviation.  Knowing mean and standard deviation you can write down an exact mathematical formula for the bell curve.  If mean is mu and standard deviation is sigma, the unique normal distribution is denoted N(mu,sigma).  We talked about the mean, median, and mode of a density curve: for a bell curve these three things coincide at the peak.  At one standard deviation from the mean of a bell curve you will find the inflection points where the curve goes from smiling to frowning or vice-versa (remember: bell curve means normal distribution, this doesn't work for other distributions).  Another property of a normal distribution: if you do a linear transformation and the old variable is normal, the new one is normal, too.  If the new variable is normal then the old variable must have also been normal.  Another property of normal distribution is the 68-95-99.7 Rule, the percentage of area falling one, two, or three standard deviations from the mean.  We talked about pseudo-random numbers, simulating pseudo-random number in StatCrunch, and the concept of seed, including fixed seed and dynamic seed.
 +
* Homework: [[Media:Stat202_2015S_HW6.pdf|Homework 6]] was assigned today with a due date of September 21, 2016.
 +
* Next class: We will review all of the above, introduce QQ plots (also called normal quantile plots), and work on homeworks 6, 7, 8.  This will complete our coverage of chapter 1, and complete our coverage of the material on exam 1.  Next week we will review, and may start new material.  Our exam will be Thursday September 22, 2016.

Revision as of 14:00, 15 September 2016

August 29, 2016

August 31, 2016

September 1, 2016

  • Today, after a review, we covered bar graphs, pie charts and stemplots, and histograms for visualizing distributions.
  • Homework: I passed out and we worked on Homework 1, due September 8, 2016.
  • Homework: I passed out and we worked on Homework 2, due September 8, 2016.
  • Homework: I passed out but did not yet assign Homework 3. Due date to be added later.
  • Reading: Moore, McCabe, Craig, pp 1-25.
  • Next class: We will cover Exploratory data analysis from The Data Professor's Guide to Basic Statistics. Homework 3 will be assigned, but instead of working on the histogram homework in class we will do a laboratory exercise on a diamonds data set. Finally, if there is time, we will proceed to talk about summary statistics (mean, median, quartiles, percentiles, 5 number summary, standard deviation) and the box plot and modified box plot.

September 7, 2016

  • Today we reviewed the concept of histograms with the call center data set. We talked about the axes of the histogram: the x-axis is the range of the values of the quantitative variable whose distribution is being visualized. This range can be restricted by adjusting the "where" input in StatCrunch. There are 3 choices for the y-axis in StatCrunch: frequency (the count of observations in each bin), the relative frequency (the proportion of observations in each bin) and the density. The point of the density was to have a vertical scale which is independent of the number of observations and bin width. We discussed exploratory data analysis and worked together to analyze the diamonds data set. Finally I introduced mean and median.
  • Homework: Homework 1 is due next class, September 8, 2016.
  • Homework: Homework 2 is due next class, September 8, 2016.
  • Homework: Homework 3 is now assigned with a due date of September 14, 2016.
  • Practice Problems: Practice Problems for Week 1.
  • Solutions to Practice Problems: Solutions to Practice Problems for Week 1.
  • Reading: Moore, McCabe & Craig, pp. 30-36.
  • Next class: We will discuss skewed and symmetric distributions, tails, center and spread, unimodal, multimodal, and bimodal distributions. We also will continue to discus mean and median, and introduce quartiles, and percentiles. We will discuss the distinction between resistant to outliers versus sensitive to outliers. We will continue our tour of summary statistics with the 5-number summary and the related box plot and modified box plots, we will pass out homework 4 and 5, talk about the sample standard deviation, and transformations. Then, if there is time we will proceed to talk about sampling -- to understand Bessel's correction in the definition of the sample standard deviation, but also because it is one of the primary course objectives.

September 8, 2016

  • Today, we discussed skewed and symmetric distributions, tails, center and spread, unimodal, multimodal, and bimodal distributions. We also discussed mean and median, quantiles, and percentiles, resistant to outliers versus sensitive to outliers. We also continued our tour of summary statistics with the 5-number summary and the related box plot and modified box plots.
  • Homework: Homework 3 was already assigned with a due date of September 14, 2016.
  • Homework: Homework 4 was assigned today with a due date of September 15, 2016.
  • Reading: Moore, McCabe & Craig, pp. 37-42.
  • Next class: We will quickly review measures of center and spread, 5 number summary, box plot and modified box plot. We will cover transformations. We will have class time for Homework 5. We will cover standard deviation and start to study sampling.

September 12, 2016

  • Today, we quickly reviewed measures of center and spread, 5 number summary, box plot and modified box plot. We covered transformations. We covered standard deviation. We covered Sampling in The Data Professor's Guide to Basic Statistics. We also discussed the definition of the number "N choose n."
  • Homework: Homework 5 was assigned today with a due date of September 21, 2016.
  • Next class: we will discuss density curves and start probability.

September 14, 2016

  • Today we reviewed standard deviation, transformations, density curves, the relationship between density curves and histograms. We discussed the fact that all density curves describe a distribution, but to be a density curve: necessarily the area under the curve must be 1 and whole density curve must be on or above the vertical axis. We talked about bell curves and the relationship between bell curves and the normal distribution (bell curves are the density curves for normal distributions). Bell curves have a very specific shape that depends on only two parameters: mean and standard deviation. Knowing mean and standard deviation you can write down an exact mathematical formula for the bell curve. If mean is mu and standard deviation is sigma, the unique normal distribution is denoted N(mu,sigma). We talked about the mean, median, and mode of a density curve: for a bell curve these three things coincide at the peak. At one standard deviation from the mean of a bell curve you will find the inflection points where the curve goes from smiling to frowning or vice-versa (remember: bell curve means normal distribution, this doesn't work for other distributions). Another property of a normal distribution: if you do a linear transformation and the old variable is normal, the new one is normal, too. If the new variable is normal then the old variable must have also been normal. Another property of normal distribution is the 68-95-99.7 Rule, the percentage of area falling one, two, or three standard deviations from the mean. We talked about pseudo-random numbers, simulating pseudo-random number in StatCrunch, and the concept of seed, including fixed seed and dynamic seed.
  • Homework: Homework 6 was assigned today with a due date of September 21, 2016.
  • Next class: We will review all of the above, introduce QQ plots (also called normal quantile plots), and work on homeworks 6, 7, 8. This will complete our coverage of chapter 1, and complete our coverage of the material on exam 1. Next week we will review, and may start new material. Our exam will be Thursday September 22, 2016.