Central Limit Theorem
- Experiment N° 0
- Link: here
Design
Empirical validation of CLT for different distributions.
section-I
import numpy as np
import matplotlib.pyplot as plt
import os
os.getcwd()
'C:\\Users\\juani\\Documents\\3_My_Jupiter_Notebooks\\4_Series_de_Tiempo\\0_2024\\exps_&_reps'
import warnings
warnings.simplefilter(action='ignore', category= FutureWarning)
# general settings
class CFG:
data_folder = '../input/tsdata-1/'
img_dim1 = 20
img_dim2 = 10
style = 'fivethirtyeight'
# adjust the parameters for displayed figures
plt.rcParams.update({'figure.figsize': (CFG.img_dim1,CFG.img_dim2)})
plt.style.use(CFG.style)
import warnings
warnings.simplefilter(action='ignore', category= FutureWarning)
# general settings
class CFG:
data_folder = '../input/tsdata-1/'
img_dim1 = 20
img_dim2 = 10
style = 'fivethirtyeight'
# adjust the parameters for displayed figures
plt.rcParams.update({'figure.figsize': (CFG.img_dim1,CFG.img_dim2)})
plt.style.use(CFG.style)
section-II
Normal Distribution
np.random.seed(42)
population_size = 10000
mu, sigma = 0, 100 # mean and standard deviation
height_population = np.random.normal(mu,sigma,population_size)
sample_size, no_of_samples = 10 , 10000
sample_means = []
for i in range(no_of_samples):
sample = np.random.choice(height_population, sample_size)
sample_mean = np.mean(sample)
sample_means.append(sample_mean)
plt.hist(height_population, bins = 100, density = True, alpha= 0.8, label = 'Population')
plt.xlabel('Height') , plt.ylabel('Frequency')
plt.title('Population and Sample Means Distribution')
plt.hist(sample_means, bins = 100, density = True, alpha= 0.8, label = 'Sample Means')
plt.legend, plt.show()

(<function matplotlib.pyplot.legend(*args, **kwargs)>, None)
Uniform Distribution
np.random.seed(42)
population_size = 10000
mini = -100
maxi = 100
height_population = np.random.uniform(mini,maxi,population_size)
sample_size, no_of_samples = 10 , 10000
sample_means = []
for i in range(no_of_samples):
sample = np.random.choice(height_population, sample_size)
sample_mean = np.mean(sample)
sample_means.append(sample_mean)
plt.hist(height_population, bins = 100, density = True, alpha= 0.8, label = 'Population')
plt.xlabel('Height') , plt.ylabel('Frequency')
plt.title('Population and Sample Means Distribution')
plt.hist(sample_means, bins = 100, density = True, alpha= 0.8, label = 'Sample Means')
plt.legend, plt.show()

(<function matplotlib.pyplot.legend(*args, **kwargs)>, None)
Beta 1 Distribution
np.random.seed(42)
mini = 8
maxi = 2
population_size = 10000
height_population = np.random.beta(mini,maxi,population_size)
sample_size, no_of_samples = 10 , 10000
sample_means = []
for i in range(no_of_samples):
sample = np.random.choice(height_population, sample_size)
sample_mean = np.mean(sample)
sample_means.append(sample_mean)
plt.hist(height_population, bins = 100, density = True, alpha= 0.8, label = 'Population')
plt.xlabel('Height') , plt.ylabel('Frequency')
plt.title('Population and Sample Means Distribution')
plt.hist(sample_means, bins = 100, density = True, alpha= 0.8, label = 'Sample Means')
plt.legend, plt.show()

(<function matplotlib.pyplot.legend(*args, **kwargs)>, None)
Beta 2 Distribution
np.random.seed(42)
mini = 2
maxi = 8
population_size = 10000
height_population = np.random.beta(mini,maxi,population_size)
sample_size, no_of_samples = 10 , 10000
sample_means = []
for i in range(no_of_samples):
sample = np.random.choice(height_population, sample_size)
sample_mean = np.mean(sample)
sample_means.append(sample_mean)
plt.hist(height_population, bins = 100, density = True, alpha= 0.8, label = 'Population')
plt.xlabel('Height') , plt.ylabel('Frequency')
plt.title('Population and Sample Means Distribution')
plt.hist(sample_means, bins = 100, density = True, alpha= 0.8, label = 'Sample Means')
plt.legend, plt.show()

(<function matplotlib.pyplot.legend(*args, **kwargs)>, None)
Beta 3 Distribution
np.random.seed(42)
mini = 0.5
maxi = 0.5
population_size = 10000
height_population = np.random.beta(mini,maxi,population_size)
sample_size, no_of_samples = 10 , 10000
sample_means = []
for i in range(no_of_samples):
sample = np.random.choice(height_population, sample_size)
sample_mean = np.mean(sample)
sample_means.append(sample_mean)
plt.hist(height_population, bins = 100, density = True, alpha= 0.8, label = 'Population')
plt.xlabel('Height') , plt.ylabel('Frequency')
plt.title('Population and Sample Means Distribution')
plt.hist(sample_means, bins = 100, density = True, alpha= 0.8, label = 'Sample Means')
plt.legend, plt.show()

(<function matplotlib.pyplot.legend(*args, **kwargs)>, None)
section-III
section-IV
Credits to Snehil Raj Choubey.