## 2018年5月21日月曜日

### Python - 数学 - データを統計量で記述する - 他のCSVデータでの実験(平均、中央値、最頻値、分散、標準偏差)

Pythonからはじめる数学入門 (Amit Saha (著)、黒川 利明 (翻訳)、オライリージャパン)の2章(データを統計量で記述する)、3.9(プログラミングチャレンジ)、問題3-3(他のCSVデータでの実験)を取り組んでみる。

コード(Emacs)

Python 3

```#!/usr/bin/env python3

from collections import Counter
import csv
import matplotlib.pyplot as plt

with open(filename) as f:
years = []
pops = []
years.append(row[0].split('-')[0])
pops.append(float(row[1]))
years.reverse()
pops.reverse()

return years, pops

with open(filename) as f:
nums = [float(line) for line in f]

return nums

def mean(nums):
s = sum(nums)
n = len(nums)
return s / n

def median(nums):
n = len(nums)
nums.sort()

if n % 2 == 0:
m1 = n / 2
m2 = n / 2 + 1
m1 = int(m1) - 1
m2 = int(m2) - 1
return (nums[m1] + nums[m2]) / 2
m = (n + 1) / 2
m = int(m) - 1
return nums[m]

def mode(nums):
c = Counter(nums)
nums_freq = c.most_common()
max_count = nums_freq[0][1]
modes = [num[0] for num in nums_freq if num[1] == max_count]
return modes

def differences(nums):
m = mean(nums)
diff = [n - m for n in nums]
return diff

def variance(nums):
diff = differences(nums)
squared_diff = [d ** 2 for d in diff]
sum_sqaured_diff = sum(squared_diff)
return sum_sqaured_diff / len(nums)

def standard_deviation(nums):
return variance(nums) ** (1 / 2)

if __name__ == '__main__':
filename = 'USA_SP_POP_TOTL.csv'
xlabels = [f'{years[i]}-{years[i+1]}' for i in range(len(years) - 1)]
diff = [pops[i + 1] - pops[i] for i in range(len(pops) - 1)]
plt.plot(xlabels, diff)
plt.xlabel('Year')
plt.ylabel('Population')
plt.xticks(range(0, len(years) - 1), rotation=20)
plt.savefig('sample3.svg')
funcs = [mean, median, mode, variance, standard_deviation]
for func in funcs:
print(f'{func.__name__.replace("_", " ")}: {func(diff)}')
```

```\$ ./sample3.py
mean: 2562366.153846154
median: 2476370.0
mode: [1945000.0, 1971000.0, 1994000.0, 2013000.0, 2033000.0, 2062000.0, 2099000.0, 2119000.0, 2128000.0, 2152000.0, 2156000.0, 2170000.0, 2198000.0, 2204000.0, 2209000.0, 2210000.0, 2235000.0, 2241000.0, 2257000.0, 2261591.0, 2320000.0, 2326224.0, 2346000.0, 2375000.0, 2414000.0, 2470000.0, 2482740.0, 2554696.0, 2609000.0, 2647000.0, 2656238.0, 2677563.0, 2697365.0, 2704000.0, 2711301.0, 2804000.0, 2806544.0, 2847000.0, 2851295.0, 2862759.0, 2863313.0, 3020000.0, 3116000.0, 3122411.0, 3152000.0, 3186000.0, 3197000.0, 3207000.0, 3263000.0, 3358000.0, 3405000.0, 3533000.0]
variance: 188985554755.28406
standard deviation: 434724.6884584358
\$
```