def acf_pacf_plot(series, lags=50, figsize=(10, 10)):
fig, axes = plt.subplots(nrows=2, ncols=1, figsize=figsize)
plotting.plot_acf(series, lags, ax=axes[0])
axes[0].set_title('Autocorrelation (ACF)')
plotting.plot_acf(series, lags, ax=axes[1], partial=True)
axes[1].set_title('Partial autocorrelation (PACF)')
def acf_pacf_plot(series, lags=50, figsize=(10,10)):
fig, axes = plt.subplots(nrows=2, ncols=1, figsize=figsize)
plotting.plot_acf(series, lags, ax=axes[0])
axes[0].set_title('Autocorrelation (ACF)')
plotting.plot_acf(series, lags, ax=axes[1], partial=True)
axes[1].set_title('Partial autocorrelation (PACF)')
return out
def simulate_76(pi=0.9, phi=0.9, v=1, T=5000):
rv = np.sqrt(v)
s = v / (1 - phi**2)
root_s = np.sqrt(s)
x = 1
out = np.empty(T)
for t in xrange(T):
flip = np.random.binomial(1, pi)
if flip:
out[t] = x = phi * x + randn() * rv
else:
out[t] = x = randn() * root_s
return out
if __name__ == '__main__':
sim = simulate_76(pi=1, T=10000)
print acf(sim, nlags=10)
plotting.plot_acf(sim)
plt.show()
yenusd = np.loadtxt('statlib/data/japan-usa1000.txt', delimiter=',')
ukusd = np.loadtxt('statlib/data/uk-usa1000.txt', delimiter=',')
return out
def simulate_76(pi=0.9, phi=0.9, v=1, T=5000):
rv = np.sqrt(v)
s = v / (1 - phi**2)
root_s = np.sqrt(s)
x = 1
out = np.empty(T)
for t in xrange(T):
flip = np.random.binomial(1, pi)
if flip:
out[t] = x = phi * x + randn() * rv
else:
out[t] = x = randn() * root_s
return out
if __name__ == '__main__':
sim = simulate_76(pi=1, T=10000)
print acf(sim, nlags=10)
plotting.plot_acf(sim)
plt.show()
yenusd = np.loadtxt('statlib/data/japan-usa1000.txt', delimiter=',')
ukusd = np.loadtxt('statlib/data/uk-usa1000.txt', delimiter=',')
def plot_acf(self, lags=50, partial=True):
plotting.plot_acf(self.data, lags, partial=partial)
def plot_acf(self, lags=50, partial=True):
plotting.plot_acf(self.data, lags, partial=partial)