| 123456789101112131415161718192021222324252627282930313233343536373839404142 |
- import pickle
- import pandas as pd
- import quandl
- import matplotlib.pyplot as plt
- from matplotlib import style
- style.use("seaborn")
- quandl.ApiConfig.api_key = "rFsSehe51RLzREtYhLfo"
- def mortgage_30yr():
- df = quandl.get("FMAC/MORTG")
- df = df[df.index > "1974-12-01"]
- df = (df["Value"] - df["Value"][0]) / df["Value"][0] * 100
- df = df.resample("M").mean()
- return df
- ax1 = plt.subplot(2, 1, 1)
- ax2 = plt.subplot(2, 1, 2, sharex=ax1)
- # initial_state_data()
- pickle_in = open("fifty_states_pct.pickle", "rb")
- HPI_data = pickle.load(pickle_in)
- # HPI_Benchmark()
- pickle_in = open("us_pct.pickle", "rb")
- benchmark = pickle.load(pickle_in)
- m30 = mortgage_30yr()
- HPI_Bench = benchmark
- state_HPI_M30 = HPI_data.join(m30)
- state_HPI_M30.rename({"Value": "M30"}, inplace=True)
- print(state_HPI_M30.corr().describe()["Value"])
|
