def _plot(self,
x,
y,
labels=None,
x_label='Return [%]',
y_label='-',
title='Backtesting results',
plot_title='',
xlim=None,
ylim=None,
pres_name=None):
if pres_name is None:
pres_name = self.presname
pres = Presenter(pres_name)
fig = plt.figure()
plt.scatter(x, y)
if xlim is not None:
plt.xlim(xlim)
if ylim is not None:
plt.ylim(ylim)
plt.xlabel(x_label)
plt.ylabel(y_label)
plt.title(plot_title)
plt.grid()
img_path = 'temp.png'
plt.savefig(img_path)
plt.close(fig)
pres.add_picture_slide(img_path, title=title)
pres.save_presentation(pres_name)
def _cm_plot(self,
x,
y,
labels=None,
x_label='Transaction Value [monetary unit]',
y_label='Transaction Costs [monetary unit]',
title='Transaction costs',
plot_title=''):
pres = Presenter(self.presname)
fig = plt.figure()
if isinstance(y[0], np.ndarray) | isinstance(y[0], list):
for i in range(0, len(y)):
plt.plot(x, y[i], label=labels[i])
plt.legend(loc='upper left')
else:
plt.plot(x, y, 'b', label=labels)
if labels is not None:
plt.legend(loc='upper left')
plt.xlabel(x_label)
plt.ylabel(y_label)
plt.title(plot_title)
img_path = 'temp.png'
plt.savefig(img_path)
plt.close(fig)
pres.add_picture_slide(img_path, title=title)
pres.save_presentation(self.presname)
os.remove(img_path)
def compare_to_moving_average(self, window):
if not os.path.exists('Data_Checks'):
os.makedirs('Data_Checks')
location = 'Data_Checks'
pres = Presenter()
pres.start_presentation(
title=''.join(['Data Checks ', self.ticker]),
subtitle=dt.datetime.today().date().isoformat())
mov_df = pd.DataFrame(index=[],
columns=[
'id', 'price_date', 'item', 'value',
'lower_limit', 'upper_limit'
])
for name in self.orig_names:
mov_avg = pd.rolling_mean(self.unchecked_data[name],
window=window,
min_periods=1)
mov_std = pd.rolling_std(self.unchecked_data[name],
window=window,
min_periods=1)
num_std = 3
# Determine range
up_lim = mov_avg + num_std * mov_std
low_lim = mov_avg - num_std * mov_std
# Plot it
dates = pd.to_datetime(self.unchecked_data['price_date'])
fig = plt.figure()
plt.xticks(rotation=45)
plt.gcf().subplots_adjust(bottom=0.15)
plt.plot(dates.values,
self.unchecked_data[name],
'r',
label='Raw data')
plt.plot(dates.values, mov_avg, 'b', label='Moving average')
plt.plot(dates.values,
up_lim,
'g',
label=''.join([str(num_std), 'std']))
plt.plot(dates.values, low_lim, 'g')
plt.xlabel('Time')
plt.ylabel('Value')
plt.legend(loc='upper left')
img_path = ''.join(['temp', self.ticker, '.png'])
plt.savefig(img_path)
plt.close(fig)
pres.add_picture_slide(img_path, title=name)
# Find all values that lie outside of this range
extr = self.unchecked_data.loc[
(self.unchecked_data[name] > up_lim) |
(self.unchecked_data[name] < low_lim)]
# Add to list
extr_select = extr.ix[:, ['id', 'price_date', name]]
extr_select.loc[:, 'item'] = [name] * len(extr_select)
extr_select.loc[:, 'upper_limit'] = up_lim[extr_select.index]
extr_select.loc[:, 'lower_limit'] = low_lim[extr_select.index]
extr_select.rename(columns={name: 'value'}, inplace=True)
mov_df = pd.concat(
[mov_df,
extr_select.reindex_axis(mov_df.columns, axis=1)])
# Print list
mov_df.to_csv(''.join([
location, '/', self.ticker, '_', self.symbol_id,
'_moving_avg_data.csv'
]),
index=False)
pres.save_presentation(''.join(
[location, '/', self.ticker, '_', self.symbol_id, '_plots.pptx']))
os.remove(img_path)
return mov_df