def update(self, dates):
"""Update alpha performance metrics for the **same** day after market close."""
dnames = self.db.alpha.distinct('dname')
si, ei = map(self.dates.index, [dates[0], dates[-1]])
sywg_returns = quote.fetch('returns', self.dates[si-20], self.dates[ei])
cnt = 0
for dname in dnames:
if self.options['alphas'] and dname not in self.options['alphas']:
continue
cursor = self.db.alpha.find(
{'dname': dname, 'date': {'$gte': self.dates[si-20], '$lte': self.dates[ei]}},
{'_id': 0, 'dvalue': 1, 'date': 1})
alpha = pd.DataFrame({row['date']: row['dvalue'] for row in cursor}).T
if len(alpha) == 0:
continue
analyser = Analyser(api.scale(api.neutralize(alpha)), data=sywg_returns)
for date in dates:
if date not in alpha.index:
continue
key = {'alpha': dname, 'date': date}
metrics = self.get_metrics(analyser, date)
self.collection.update(key, {'$set': metrics}, upsert=True)
cnt += 1
if len(dates) == 1:
self.logger.info('UPSERT documents for {} alphas into (c: [{}]) of (d: [{}]) on {}',
cnt, self.collection.name, self.db.name, dates[0])
else:
self.logger.info('UPSERT documents for {} alphas into (c: [{}]) of (d: [{}]) from {} to {}',
cnt, self.collection.name, self.db.name, dates[0], dates[-1])
def __init__(self, alpha, n, rank=None):
self.alpha = api.format(alpha)
self.rank_alpha = self.alpha.rank(axis=1, ascending=False)
self.rank_alpha = self.rank_alpha[self.rank_alpha <= n]
if rank is None:
self.alpha = (self.rank_alpha <= n).astype(float)
else:
if rank < 0:
self.alpha = self.alpha[self.rank_alpha <= n]
else:
self.alpha = rank - np.floor(self.rank_alpha / (n + 1) * rank)
self.alpha = api.scale(self.alpha)
self.dates = dateutil.to_datestr(self.alpha.index)
def __init__(self, alpha=None, data=None, index_data=None):
if alpha is not None:
self.alpha = api.scale(alpha)
self.intervals = self.alpha.index.date
self.IC, self.rIC = {}, {}
self.AC, self.rAC = {}, {}
self.turnover = None
self.returns = None
if data is not None:
self.data = data.ix[self.alpha.index]
if index_data is not None:
self.index_data = index_data.ix[self.alpha.index]
def __init__(self, alpha, data, index_data=None):
self.alpha = api.scale(alpha)
self.dates = np.unique(self.alpha.index.date)
self.intervals = len(self.alpha) / len(self.dates)
self.IC_t, self.rIC_t = None, None
self.IC_h, self.rIC_h = None, None
self.IC_d, self.rIC_d = None, None
self.AC_t, self.rAC_t = None, None
self.AC_h, self.rAC_h = None, None
self.AC_d, self.rAC_d = None, None
self.turnover_t, self.turnover_h, self.turnover_d = None, None, None
self.turnover, self.returns = None, None
self.data = data.ix[self.alpha.index]
if index_data is not None:
self.index_data = index_data.ix[self.alpha.index]
if args.alpha:
if not args.alphas:
args.alphas = []
if args.alpha not in args.alphas:
args.alphas.append(args.alpha)
assert args.alphas
alphas, dates = {}, None
for alpha in args.alphas:
try:
alphadf = read_frame(alpha, args.ftype)
if dates is None:
dates = alphadf.index
else:
dates = dates.union(alphadf.index)
alphas[alpha] = api.scale(api.neutralize(alphadf))
except:
print '[WARNING] Failed to parse file', alpha
sywg_returns = sywgquote_fetcher.fetch('returns',
dates[0].strftime('%Y%m%d'),
dates[-1].strftime('%Y%m%d'),
level=args.level)
for alpha, alphadf in alphas.iteritems():
analyser = Analyser(alphadf, data=sywg_returns)
alphas[alpha] = analyser
pool = multiprocessing.Pool(args.threads)
res = pool.imap_unordered(
worker, ((alpha, analyser, args.cost, args.by, args.group, args.freq)
for alpha, analyser in alphas.iteritems()))
if args.alpha:
if not args.alphas:
args.alphas = []
if args.alpha not in args.alphas:
args.alphas.append(args.alpha)
assert args.alphas
alphas, dates = {}, None
for alpha in args.alphas:
try:
alphadf = read_frame(alpha, args.ftype)
if dates is None:
dates = alphadf.index
else:
dates = dates.union(alphadf.index)
alphas[alpha] = api.scale(api.neutralize(alphadf))
except:
print '[WARNING] Failed to parse file', alpha
sywg_returns = sywgquote_fetcher.fetch('returns', dates[0].strftime('%Y%m%d'), dates[-1].strftime('%Y%m%d'), level=args.level)
for alpha, alphadf in alphas.iteritems():
analyser = Analyser(alphadf, data=sywg_returns)
alphas[alpha] = analyser
pool = multiprocessing.Pool(args.threads)
res = pool.imap_unordered(worker, ((alpha, analyser, args.cost, args.by, args.group, args.freq) for alpha, analyser in alphas.iteritems()))
pool.close()
pool.join()
for alpha, summary in res:
print alpha
def test_scale(self):
df = pd.DataFrame(np.random.randn(20, 40))
df = api.scale(df)
self.assertTrue(np.allclose(np.abs(df).sum(axis=1), 1))