class Beta(CustomFactor):
outputs = ['beta', 'residual_var']
inputs = [DailyReturns(), DailyReturns()[symbol('SPY')]]
window_length = 252
params = ('standardize',)
def compute(self, today, assets, out, assets_returns, market_returns, standardize):
allowed_missing_percentage = 0.25
allowed_missing_count = int(allowed_missing_percentage * self.window_length)
(out.beta, out.residual_var) = beta_residual(assets_returns, market_returns, allowed_missing_count, standardize)
class MarketDispersion(CustomFactor):
inputs = [DailyReturns()]
window_length = 1
window_safe = True
def compute(self, today, assets, out, returns):
# returns are days in rows, assets across columns
out[:] = np.sqrt(np.nanmean((returns - np.nanmean(returns))**2))
class MarketVolatility(CustomFactor):
inputs = [DailyReturns()]
window_length = 1
window_safe = True
def compute(self, today, assets, out, returns):
mkt_returns = np.nanmean(returns, axis=1)
out[:] = np.sqrt(260. * np.nanmean(
(mkt_returns - np.nanmean(mkt_returns))**2))
class MarketVolatility(CustomFactor):
inputs = [DailyReturns()]
window_length = 1 # We'll want to set this in the constructor when creating the object.
window_safe = True
def compute(self, today, assets, out, returns):
DAILY_TO_ANNUAL_SCALAR = 252. # 252 trading days in a year
"""
For each row (each row represents one day of returns),
calculate the average of the cross-section of stock returns
So that market_returns has one value for each day in the window_length
So choose the appropriate axis (please see hints above)
"""
mkt_returns = np.nanmean(returns, axis=1)
"""
Calculate the mean of market returns
"""
mkt_returns_mu = np.nanmean(mkt_returns)
"""
Calculate the standard deviation of the market returns, then annualize them.
"""
out[:] = np.sqrt(DAILY_TO_ANNUAL_SCALAR * np.nanmean(
(mkt_returns - mkt_returns_mu)**2))
def test_daily_returns_is_special_case_of_returns(self):
self.check_equivalent_terms({
'daily': DailyReturns(),
'manual_daily': Returns(window_length=2),
})