Professional Documents
Culture Documents
2. Tell EXCEL where the data are the ozone data are in ozone2, check labels, new worksheet, Chart Output, and we will see 3. the output is
Bin 1 1.452 1.904 2.356 2.808 3.26 3.712 4.164 4.616 5.068 More Frequency 1 0 2 16 22 19 19 9 16 5 2
Histogram
Frequency 30 20 10 0 3.26
3.712
1.452
1.904
2.356
2.808
4.164
4.616
5.068
Bin
Which, of course, doesnt look good. We will fix the looks later. But, first 4. A frequency distribution is a grouping of a data sets values into classes or bins called class intervals. The classes should be of equal width and should be non-overlapping. The number of bins is subjective (Yogi Berra Principle). There are some rules of thumb (ROT) ROT 1: size of data set number of bins ROT 2: number of bins < 100 5-10 = 102 200 11-15 >200 13-20 Without being told EXCEL chooses 11 bins, which satisfies ROT 1 and ROT 2. Let W denote the bin width. Then EXCELs W is 0.452 and EXCELs frequency distribution is on the overhead. 5. We could argue the minimum ozone is 1.00 on day 17 and the maximum ozone is 5.52 on day 77. This means the Range = R = 5.52 1.00 = 4.52, which may be why EXCELs W is .452. We can argue if there are 11 bins then to cover all the data points in the data set W > 4.52 / 11 = .41090909 (as EXCEL does). Any W greater than .41090909 will work. I like .5 which means the number of bins is 10. This is ok. How do we tell EXCEL? The following will do it, in ozone1 and sheet1. We make a separate column for the upper values of the bins.
More
Frequency
After telling EXCEL these are the bins we want the output is now
bin 1.25 1.75 2.25 2.75 3.25 3.75 4.25 4.75 5.25 5.75 More
Frequency 1 1 11 23 24 20 12 12 6 1 0
Histogram
Frequency 30 20 10 0
4.25
1.25
1.75
2.25
2.75
3.25
3.75
4.75
5.25
5.75
bin
and the frequency distribution is now greater than less than or equal 0.75 1.25 1.25 1.75 1.75 2.25 2.25 2.75 2.75 3.25 3.25 3.75 3.75 4.25 4.25 4.75 4.75 5.25 5.25 5.75
frequency 1 1 11 23 24 20 12 12 6 1
6. That leaves the issue of fixing the looks of the plot double click on a column, set the gaps to zero, and see the overhead.
More
Frequency