def test_plot_kwarg(self):
# Check with the matplotlib.pyplot module
fig = plt.figure()
fig.add_subplot(111)
stats.boxcox_normplot(self.x, -20, 20, plot=plt)
plt.close()
# Check that a Matplotlib Axes object is accepted
fig.add_subplot(111)
ax = fig.add_subplot(111)
stats.boxcox_normplot(self.x, -20, 20, plot=ax)
plt.close()
def test_plot_kwarg(self):
# Check with the matplotlib.pyplot module
fig = plt.figure()
fig.add_subplot(111)
stats.boxcox_normplot(self.x, -20, 20, plot=plt)
plt.close()
# Check that a Matplotlib Axes object is accepted
fig.add_subplot(111)
ax = fig.add_subplot(111)
stats.boxcox_normplot(self.x, -20, 20, plot=ax)
plt.close()
def test(samples, la=-20, lb=20):
fig = plt.figure()
ax = fig.add_subplot(111)
prob = boxcox_normplot(samples, la, lb, plot=ax)
best_lambda = boxcox_normmax(samples)
ax.axvline(best_lambda, color='r')
plt.show()
def test_basic(self):
N = 5
lmbdas, ppcc = stats.boxcox_normplot(self.x, -10, 10, N=N)
ppcc_expected = [0.57783375, 0.83610988, 0.97524311, 0.99756057,
0.95843297]
assert_allclose(lmbdas, np.linspace(-10, 10, num=N))
assert_allclose(ppcc, ppcc_expected)
def train_theta_boxcox(ts, seasonality, n):
theta_bc = Theta(theta=0, season_mode=SeasonalityMode.NONE)
shiftdata = 0
if (ts.univariate_values() < 0).any():
shiftdata = -ts.min() + 100
ts = ts + shiftdata
new_values, lmbd = boxcox(ts.univariate_values())
if lmbd < 0:
lmbds, value = boxcox_normplot(ts.univariate_values(),
lmbd - 1,
0,
N=100)
if np.isclose(value[0], 0):
lmbd = lmbds[np.argmax(value)]
new_values = boxcox(ts.univariate_values(), lmbd)
if np.isclose(new_values, new_values[0]).all():
lmbd = 0
new_values = boxcox(ts.univariate_values(), lmbd)
ts = TimeSeries.from_times_and_values(ts.time_index(), new_values)
theta_bc.fit(ts)
forecast = theta_bc.predict(n)
new_values = inv_boxcox(forecast.univariate_values(), lmbd)
forecast = TimeSeries.from_times_and_values(seasonality.time_index(),
new_values)
if shiftdata > 0:
forecast = forecast - shiftdata
forecast = forecast * seasonality
if (forecast.univariate_values() < 0).any():
indices = seasonality.time_index()[forecast < 0]
forecast = forecast.update(indices,
np.zeros(len(indices)),
inplace=True)
return forecast
def box_cox_normality_plot(series, lambda_min=-2, lambda_max=2, N=100,
ax=None, show=True, save=False):
if ax is None:
fig, ax = plt.subplots()
lambdas, corrs = stats.boxcox_normplot(series, lambda_min,
lambda_max, plot=ax, N=N)
max_corr_value = corrs.max()
max_corr_lambda = lambdas[corrs.argmax()]
show_and_save_plot(show=show, save=save,
filename="box_cox_normality.png")
return lambdas, corrs, max_corr_lambda, max_corr_value
def histogram_boxcox_plot(var_name, var_data):
shift = np.amin(var_data)
shift = min([shift, 0])
shift_output_var = [x + shift for x in var_data]
#Histogram and boxcox plot for output
fig = plt.figure(figsize=(8,11), dpi=300)
ax1 = fig.add_subplot(211)
ax2 = fig.add_subplot(212)
#histogram of original data
ax1.set_title("Distribution of " + var_name)
ax1.set_xlabel(var_name)
var_data.hist(ax=ax1, rasterized=True)
#boxcox norm plot of shifted data
lmbdas, ppcc = stats.boxcox_normplot(shift_output_var, -10, 10, plot=ax2)
ax2.plot(lmbdas, ppcc, 'bo')
shift_output_var_t, maxlog = stats.boxcox(shift_output_var)
boxcox_output_var_t = [x - shift for x in shift_output_var_t]
#adding vertical line to plot
ax2.axvline(maxlog, color='r')
ax2.text(maxlog + 0.1, 0, s=str(maxlog))
return fig
def test_empty(self):
assert_(stats.boxcox_normplot([], 0, 1).size == 0)
def test_empty(self):
assert_(stats.boxcox_normplot([], 0, 1).size == 0)
from scipy import stats import matplotlib.pyplot as plt # Generate some non-normally distributed data, and create a Box-Cox plot: x = stats.loggamma.rvs(5, size=500) + 5 fig = plt.figure() ax = fig.add_subplot(111) prob = stats.boxcox_normplot(x, -20, 20, plot=ax) # Determine and plot the optimal ``lmbda`` to transform ``x`` and plot it in # the same plot: _, maxlog = stats.boxcox(x) ax.axvline(maxlog, color='r') plt.show()
# First we generate some random data from a Tukey-Lambda distribution, # with shape parameter -0.7: from scipy import stats import matplotlib.pyplot as plt np.random.seed(1234567) x = stats.tukeylambda.rvs(-0.7, loc=2, scale=0.5, size=10000) + 1e4 # Now we explore this data with a PPCC plot as well as the related # probability plot and Box-Cox normplot. A red line is drawn where we # expect the PPCC value to be maximal (at the shape parameter -0.7 used # above): fig = plt.figure(figsize=(12, 4)) ax1 = fig.add_subplot(131) ax2 = fig.add_subplot(132) ax3 = fig.add_subplot(133) res = stats.probplot(x, plot=ax1) res = stats.boxcox_normplot(x, -5, 5, plot=ax2) res = stats.ppcc_plot(x, -5, 5, plot=ax3) ax3.vlines(-0.7, 0, 1, colors='r', label='Expected shape value') plt.show()
from scipy import stats import matplotlib.pyplot as plt np.random.seed(1234) # make this example reproducible # Generate some data and determine optimal ``lmbda`` in various ways: x = stats.loggamma.rvs(5, size=30) + 5 y, lmax_mle = stats.boxcox(x) lmax_pearsonr = stats.boxcox_normmax(x) lmax_mle # 7.177... lmax_pearsonr # 7.916... stats.boxcox_normmax(x, method='all') # array([ 7.91667384, 7.17718692]) fig = plt.figure() ax = fig.add_subplot(111) prob = stats.boxcox_normplot(x, -10, 10, plot=ax) ax.axvline(lmax_mle, color='r') ax.axvline(lmax_pearsonr, color='g', ls='--') plt.show()
plt.show()
# inspect histogram
Newton_wind_temp['Daily_Rainfall(mm)'].hist(bins=30)
plt.title('Histogram of Dependent Variable')
plt.show()
# diagnose/inspect residual normality using qqplot:
stats.probplot(Newton_wind_temp['resid'], dist="norm", plot=plt)
plt.figure(figsize=(8,8))
plt.show()
#%%
# inspect sqrt histogram
np.sqrt(Newton_wind_temp['Daily_Rainfall(mm)']).hist(bins=30)
plt.title('Histogram of Dependent Variable')
plt.show()
# inspect log histogram
np.log(Newton_wind_temp['Daily_Rainfall(mm)']+1).hist(bins=30)
plt.title('Histogram of Dependent Variable')
plt.show()
# inspect to examine for box cox transformation
fig = plt.figure()
ax = fig.add_subplot(111)
prob = stats.boxcox_normplot(Newton_wind_temp['Daily_Rainfall(mm)']+1, 1, 3, plot=ax)
#%%