import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

n = 1000000
data = pd.DataFrame(
  {
      'A' : np.random.randn(n),
      'B' : 1.5 + 2.5 * np.random.randn(n),
      'C' : np.random.uniform(5,32,n)
  }
)

data.describe()

plt.hist(data['A'])
plt.hist(data['B'])
plt.hist(data['C'])


new_data = pd.DataFrame(
  {
      'Column Name': column_names,
      'A' : np.random.randn(a),
      'B' : 1.5 + 2.5 * np.random.randn(a),
      'C' : np.random.uniform(5,32,a)
  }
  # si queremos empalmar un data set existente.
  #  , index = range(42, 42 + a)

)