Distribution of Dollars Spent at Chipotle per Order

Chipotle order distribution

import codecs
import json
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from scipy import stats
from dateutil import *
import math
import statsmodels.formula.api as sm
from collections import Counter
import csv


# DATA FROM https://github.com/TheUpshot/chipotle
orders = []
with open("orders.tsv") as tsv:
    for line in csv.reader(tsv, dialect="excel-tab"):
        orders.append(line)
columns = orders.pop(0)
        
chipotle_df=pd.DataFrame(orders)
chipotle_df.columns = columns

def price_to_float(price):
    stripped = price.replace("$","")
    try:
        return float(stripped)
    except:
        return None

chipotle_df['dollars']=chipotle_df['item_price'].map(price_to_float)
order_gb = chipotle_df.groupby("order_id")
order_df = order_gb.aggregate(np.sum)
order_gb['dollars'].sum().hist(bins=40, normed=True, color=sns.xkcd_rgb["light blue"])
order_gb['dollars'].sum().plot(kind="kde", color=sns.xkcd_rgb["pale red"])
plt.title("Histogram and k density of dollars spent per order at Chipotle n = 1834")
plt.show()

import scipy.stats as stats
stats.probplot(order_df['dollars'], dist="norm", plot=pylab)
plt.title("QQ Plot of dollars spent per order at Chipotle, n = 1834")
pylab.show()