Chapter 9: Hotspot Analysis

Activity 21: Estimating Density in One Dimension

Solution:

1. Open a new notebook and install all the necessary libraries.

   get_ipython().run_line_magic('matplotlib', 'inline')
   import matplotlib.pyplot as plt
   import numpy
   import pandas
   import seaborn
   import sklearn.datasets
   import sklearn.model_selection
   import sklearn.neighbors
   seaborn.set()

2. Sample 1,000 data points from the standard normal distribution. Add 3.5 to each of the last 625 values of the sample (that is, the indices between 375 and 1,000). To do this, set a random state of 100 using numpy.random.RandomState to guarantee the same sampled values, and then randomly generate the data points using the randn(1000) call:

   rand = numpy.random.RandomState(100)
   vals = rand.randn(1000)  # standard normal
   vals[375:] += 3.5

3. Plot the 1,000-point sample data as a histogram and add a scatterplot below it:

   fig, ax = plt.subplots(figsize=(14, 10))
   ax.hist(vals, bins=50, density...