TripCount

trips.py

from scipy.stats import poisson
import matplotlib.pyplot as plt


import numpy as np


fig, ax = plt.subplots(1, 1)

x = np.fromfile('tripcountssample.txt',
                           sep='\n', dtype=np.int64)

mu = np.mean(x)


mean, var, skew, kurt = poisson.stats(mu, moments='mvsk')


print ("Mean, Variance, Skew, Kurtosis",mean, var , skew , kurt )


def plot_cdf(data, plot_range=None, scale_to=None, **kwargs):
    num_bins= len(data)
    sorted_data = np.array(sorted(data), dtype=np.float64)
    data_range = sorted_data[-1] - sorted_data[0]
    counts, bin_edges = np.histogram(sorted_data, bins=num_bins)
    xvalues = bin_edges[1:]
    yvalues = np.cumsum(counts)
    if plot_range is None:
        xmin = xvalues[0]
        xmax = xvalues[-1]
    else:
        xmin, xmax = plot_range
    # pad the arrays
    xvalues = np.concatenate([[xmin, xvalues[0]], xvalues, [xmax]])
    yvalues = np.concatenate([[0.0, 0.0],        yvalues, [yvalues.max()]])
    if scale_to:
        yvalues = yvalues / len(data) * scale_to
    plt.axis([xmin, xmax, 0, yvalues.max()])
    return plt.step(xvalues, yvalues, **kwargs)


# Plot the graph and look at results.

plot_cdf(x, plot_range=[30, 60],
            scale_to=100, lw=2)

plt.grid(lw=1)
plt.xlabel('Counts (x.1mm)', fontsize=18)
plt.ylabel('Percent', fontsize=18)

plt.show()


# How do we predict using cumulative distribution taking counts 36

predict  = poisson.cdf(36,mu)

print ("The precentage prediction for trip count 36 ", predict * 100)


predict  = poisson.cdf(44,mu)