Project 2 Histogram Solution

Write a C++ program that reads a list of numbers in from a file,

prints out some statistics (count, range, mean, standard deviation),

and then prints a histogram of the values in ASCII.

The histogram should span between the min and max of the values,

and have 20 evenly spaced bins (for now).




The filename should be given on the command-line.

Assume it has one value on each line.




Do the calculation of the statistics yourself; don't use library functions.




Since the input file could have an arbitrary number of values, you

will have to store them in a dynamic data structure, like a vector.

An important part of this program is setting up an array or similar

data structure to count how many items are in each bin (try to do

this in one pass).




Here is an example of running on daily temperatures over 3 days in the summer for BCS.

hourly weather data: https://w1.weather.gov/data/obhistory/KCLL.html




histo wx.dat

num_vals: 72

mean: 84.6806

std_dev: 5.91097

range: 77 - 95

77.00 - 77.90: ****

77.90 - 78.80: ************

78.80 - 79.70: ***

79.70 - 80.60: ***

80.60 - 81.50: ********

81.50 - 82.40: *****

82.40 - 83.30: **

83.30 - 84.20: **

84.20 - 85.10: ***

85.10 - 86.00:

86.00 - 86.90: *

86.90 - 87.80: ****

87.80 - 88.70: *

88.70 - 89.60: ******

89.60 - 90.50: **

90.50 - 91.40: ***

91.40 - 92.30: **

92.30 - 93.20: ***

93.20 - 94.10: ***

94.10 - 95.00: *****







Get your own data:




examples:

daily temperature in BCS in a given month

aggie football scores over last couple years (or all NCAA games)

stock price of Google (GOOG) over the last year

exam grade distribution for students in a class




Note: if you have a table of interesting data, you might be able to

open it as a spreadsheet delete the other columns and save the

column of interest as a text file.







--------------------------------------------------------




Optional:




1. Add a command line flag to specify the number of bins.




2. Other options might be the starting value, the bin size.




3. If the total number of observations is very large (like 10,000), you

might want to divide the counts by a scaling factor, so the output doesn't

have too many asterisks.




4. If your input file has multiple columns (separated by tab or space), then

allow the user to indicate which column to read.