Normal Distribution is one hump and symmetricalOne standard deviation: 0.68Two standard deviations: 0.95Three standard deviations: 0.997Use the pnorm function to find cumulative proportions of a distribution that's to the left of a specific number: pnorm(value, mean, SD)Slide 29: mom’s diet effect on premature birth: vitamins only vs. vitamins and dietary supplementOnly vitamins shows a more normal distributionSlide 32: z score is how many standard deviations a value is from the mean. Z score= value – mean/ standard deviationSlide 33: 1 SD above the mean z score of 1 and positiveBelow the mean z score that is negativeSlide 34: answer is -2. You take 59.5-64.5/2.5 and get -2.Slide 35: 68-64.5/2.5 = 1.4Slide 36: the correlation coefficient uses the z score in its formula : you take the z score of the x variable * the z score of the y variable, add them all together and divide by n-1Slide 37: a distribution of z scores will follow the N(0,1) distribution. If you take any normal distribution and convert the values into z scores and look at the distribution of the z scores, it will result in N(0,1) which means that there is a mean of 0 and a SD of 1.Slide 38-39: how do you know if it’s a normal distribution? You make a histogram of the variable, take the mean and SD of that variable and create a probability density function for a normal distribution with that mean and standard deviation. Then, you put the pdf over the histogram and compare.Slide 40: a pdf is what a histogram would look like if the distribution was totally normal. The graph on this slide is not too bad, it fits fairly well.Slide 41: the peak of the graph is to the right of the middle and this is not a normal distributionSlide 42: Answer – it’s hard to tell. You could make an argument for either B, C, or D. D was the most popular answer amongst the class.Slide 43: qq plot: if the distribution if approximately normal, the scatterplot will show a straight line of dotsSlide 44: this is a normal distribution so the qq plot is basically a straight lineSlide 45-46: this graph is less normal as shown by the qq plot. It’s not really straight so it’s not totally normally distributedSlide 47:very weird graph, the qq plot is way off and has a curved shape so it’s definitely not a normal distribution.Slide 48: this is a right skewed distributionSlide 49: answer: AThe upper Right graph is somewhat normalThe upper left is curvedThe bottom right is not normal at allNormal Distribution 03/02/2014-Normal Distribution is one hump and symmetrical-One standard deviation: 0.68-Two standard deviations: 0.95-Three standard deviations: 0.997-Use the pnorm function to find cumulative proportions of a distribution that's to the left of a specific number: pnorm(value, mean, SD)-Slide 29: mom’s diet effect on premature birth: vitamins only vs. vitamins and dietary supplement -Only vitamins shows a more normal distribution-Slide 32: z score is how many standard deviations a value is from the mean. Z score= value – mean/ standard deviation-Slide 33: 1 SD above the mean z score of 1 and positive-Below the mean z score that is negative-Slide 34: answer is -2. You take 59.5-64.5/2.5 and get -2.-Slide 35: 68-64.5/2.5 = 1.4 -Slide 36: the correlation coefficient uses the z score in its formula : you take the z score of the x variable * the z score of the y variable, add them all together and divide by n-1-Slide 37: a distribution of z scores will follow the N(0,1) distribution. If you take any normal distribution and convert the values into z scores and lookat the distribution of the z scores, it will result in N(0,1) which means that there is a mean of 0 and a SD of 1.-Slide 38-39: how do you know if it’s a normal distribution? You make a histogram of the variable, take the mean and SD of that variable and create a probability density function for a normal distribution with that mean and standard deviation. Then, you put the pdf over the histogram and compare. -Slide 40: a pdf is what a histogram would look like if the distribution was totally normal. The graph on this slide is not too bad, it fits fairly well. -Slide 41: the peak of the graph is to the right of the middle and this is not a normal distribution-Slide 42: Answer – it’s hard to tell. You could make an argument for either B, C, or D. D was the most popular answer amongst the class.-Slide 43: qq plot: if the distribution if approximately normal, the scatterplot will show a straight line of dots-Slide 44: this is a normal distribution so the qq plot is basically a straight line-Slide 45-46: this graph is less normal as shown by the qq plot. It’s not really straight so it’s not totally normally distributed-Slide 47:very weird graph, the qq plot is way off and has a curved shape so it’s definitely not a normal distribution.-Slide 48: this is a right skewed distribution -Slide 49: answer: A-The upper Right graph is somewhat normal -The upper left is curved-The bottom right is not normal at
View Full Document