Uniform and Normal distributions https://davidwills.us/math103/distro_generator.html Generate a UNIFORM distribution of 100 whole numbers from 1 to 100. uniform distribution: each number in the range is equally likely to occur. Copy and paste it to freq dist, histogram, stats *** https://davidwills.us/math103/freq_histogram.html n=____ mean=______ SD s=________ range/sqrt(12)=________ Is each quartile of the boxplot the same length:______ (may or may not be...) Make a histogram with class width 1. How many times will each number on average be in the range:___ What's the minimum # of times of this data:___ What's the maximum # of times of this data:___ i.e. there is variation Leave all that alone. Open another copy of the "freq dist, histogram, stats ***" Generate and Copy and paste a uniform distribution of 1000 whole numbers from 1 to 100. n=____ mean=______ SD s=________ range/sqrt(12)=________ basically the same values as before. Is each quartile of the boxplot the same length:______ (now we're cookin') Make a histogram with class width 1. How many times will each number on average be in the range:___ What's the minimum # of times of this data:___ What's the maximum # of times of this data:___ i.e. there is variation This up and down is the randomness to expect of a uniform distribution. If the bars are all the same height then your spidey senses should be tingling that something "special" is happening. or n is very large. As n gets larger, the histogram gets "flatter". Generate 100,000 uniformly distributed from 1 to 100. Look at the flattening "stickgram". How many datums are "statistically significant" i.e. more than 2 SD from the mean:____ "Trick question" because in a uniform distribution ALL the data is within 2 SD of the mean! ************************************************************************* Generate and Copy and paste a NORMAL distribution with mean 50 and SD 10 of 100 whole numbers. n=___ mean=______ SD s=________ median and mode near the mean?:_____ Boxplot symmetric?:____ #Outliers:____ How are the two tails' lengths compared to the IQR (i.e. the box): ___________________________ Histogram of class widths 1 and of 5 showing symmetric mound-like shape?:____ How many datums are "statistically significant" (more than 2s from the mean):______ Open another copy of the "freq dist, histogram, stats ***" Generate and Copy and paste a normal distribution with same mean and SD of 1000 whole numbers. n=___ mean=______ SD s=________ median and mode very near the mean?:_____ Boxplot symmetric?:____ #Outliers:____ How are the two tails' lengths compared to the IQR (i.e. the box): ___________________________ Histogram of class widths 1 and of 5 showing symmetric mound-like peaked shape?:____ How many datums are "statistically significant" (more than 2s from the mean):______ What percent of the data is this:_____% Empirical Rule says what percent of normal data is within 2 SD of the mean:____