기술 통계 관련 그래프

다음 그래프들은 전자책 파이썬과 함께하는 통계이야기 0, 1, 2 장에 수록된 그림들의 코드들입니다.

import numpy as np 
import pandas as pd
from scipy import stats
import matplotlib.pyplot as plt
import seaborn as sns
sns.set_style("darkgrid")

#fig 021
plt.figure(figsize=(4,3))
plt.subplot(1,2,1)
plt.arrow(0,0, 1, 2, width=0.02, color="b", label="vecor a")
plt.legend(loc="best", labelcolor="linecolor", frameon=False)
plt.title("(a)")
plt.subplot(1,2,2)
plt.arrow(0,0, 1, 2, width=0.02, color="g", label="vector a1")
plt.arrow(0,0, 2, 1, width=0.02, color="r", label="vector a2")
plt.yticks([])
plt.legend(loc="best", labelcolor="linecolor", frameon=False)
plt.title("(b)")
plt.show()

#fig 031
x=np.sort(stats.norm.rvs(loc=2, scale=np.sqrt(3), size=1000, random_state=3))
y=stats.norm.pdf(x, loc=2, scale=np.sqrt(3))
plt.figure(figsize=(5,3))
plt.hist(x, bins=10, density=True, alpha=0.2, rwidth=0.8)
plt.plot(x, y, color="r")
plt.xlabel("x", size="13")
plt.ylabel("Density", rotation="horizontal", labelpad=20, size="13")
plt.show()

#fig 131
np.random.seed(3)
x1=np.random.randint(1, 100, 100)
x=np.append(x1, 150)
q1, q2, q3=np.quantile(x, [0.25, 0.5, 0.75])
q1, q2, q3
plt.figure(figsize=(4, 3))
plt.boxplot(x)
coodi=[1, 25, 45, 67, 98, 145]
nme=['min', 'Q1', 'Q2', 'Q3', 'max', 'outlier']
for i, j in zip(coodi, nme):
  plt.text(1.1, i, j, weight="bold", color="red")
plt.show()

#fig 221
p={}
for i in range(1, 1000):
    x=np.random.randint(1, 7, size=i)
    p[i]=len(np.where(x==1)[0])/i
re=np.random.choice(list(p.values()), 4)
plt.figure(figsize=(4, 3))
plt.plot(p.keys(), p.values())
plt.hlines(1/6, 0, 1000, color="red", linestyle="--")
plt.xlabel("# of trial", size=12, weight="bold")
plt.ylabel("Probability", size=12, weight="bold")
plt.ylim(0.05)
plt.text(910, 0.2, "p=1/6", color="red" , size="11")
plt.show()

#fig 241
x=range(1, 7)
y=np.repeat(1/6, 6)
plt.figure(figsize=(3,2))
plt.scatter(x, y, s=50)
plt.hlines(1/6, 1, 6, linestyle="--", color="g")
plt.xlabel("x, number of dice")
plt.ylabel("PMF")
plt.xticks([1, 2, 3, 4, 5,6])
plt.yticks(np.linspace(0, 0.3333, 3),['0', r'$\frac{1}{6}$', r'$\frac{1}{3}$'])
plt.show()

#fig 242
from scipy import special
S=np.array([0,1,2,3, 4])
p=np.array([special.comb(4, i)*(1/2)**i*(1/2)**(4-i) for i in S])
plt.figure(figsize=(4,3))
plt.bar([0, 1, 2, 3, 4], p)
plt.xlabel("# of odd")
plt.ylabel("Prob.")
plt.show()

#fig 243
n=10
p=np.array([0.1, 0.5, 0.8])
s=np.arange(n).reshape(10,1)
pmf=stats.binom.pmf(s, n, p)
plt.figure(figsize=(10, 3))
col=['g', 'b', 'r']
lbl=["p=0.1", "p=0.5", "p=0.8"]
for i in range(3):
    plt.subplot(1,3,i+1)
    plt.bar(range(pmf.shape[0]), pmf[:,i], color=col[i], alpha=0.7, label=lbl[i])
    plt.legend(loc="best")
    plt.xticks(range(10))
    plt.xlabel("trial number")
    plt.ylim(0, 0.4)
    if i!=0:
        plt.ylabel('')
        plt.yticks([])
    else:
        plt.ylabel('probability')
plt.show()

#fig 251
x=np.linspace(-3, 3, 100)
p=stats.norm.pdf(x)
idx=np.where(x>1)[0]
fig, ax=plt.subplots(figsize=(4,3))
ax.plot(x, p, label="N(0, 1)")
ax.fill_between(x[idx], p[idx], alpha=0.5)
ax.spines['left'].set_position('center')
ax.spines['right'].set_visible(False)
ax.spines['top'].set_visible(False)
ax.spines['bottom'].set_position(('data', 0))
ax.set_yticks([])
ax.set_xticks([])
ax.text(1, -0.03, "c", fontdict={"fontsize":12, "fontweight":"bold", "color":"blue"})
ax.set_xlabel("x", loc="right")
ax.set_ylabel("pdf", loc="center")
ax.legend(loc="best")
plt.show()