def liquidate_shares(s3_client: S3Client, s3_bucket: str, trading_day: Arrow,
transactions: pd.DataFrame) -> pd.DataFrame:
"""Sells all current positions in portfolio, based on current positions."""
if transactions.empty:
return
# use ONLY the transactions dataframe to identify what stocks to liquidate
# all_positions = transactions.groupby('symbol').amount.sum()
# df = all_positions[all_positions >= 0].to_frame()
stock_positions = transactions.groupby("symbol").amount.sum()
# filter out the rows with amount = 0
df = stock_positions[stock_positions > 0].to_frame()
df["symbol"] = df.index
df["amount"] = df["amount"] * -1
df["price"] = df.apply(lambda x: qm.equities.historical.get_price(
s3_client, s3_bucket, x["symbol"], trading_day),
axis=1)
df["txn_dollars"] = df["price"] * df["amount"] * -1
df["date"] = pd.to_datetime(arrow.get(trading_day.date()).datetime)
df.set_index("date", inplace=True)
return df
def purchase_new_shares(
s3_client: S3Client,
s3_bucket: str,
trading_day: Arrow,
available_cash: float,
num_holdings: int,
weight_type: WeightType,
market_cap_percentile: int,
) -> pd.DataFrame:
"""
Purchases shares of quality momentum stocks.
Index: DatetimeIndex
Columns: Index(['amount', 'price', 'symbol', 'txn_dollars'], dtype='object')
"""
tickers = qm.algorithms.calculate_momentum.get_quality_momentum_stocks(
s3_client, s3_bucket, trading_day, num_holdings, market_cap_percentile)
if weight_type == WeightType.value_weighted:
raise NotImplementedError(
"Need to grab trading_day market cap to determine size of position"
)
data = [(x, float(math.floor(available_cash / len(tickers))))
for x in tickers]
df = pd.DataFrame(data, columns=["symbol", "position_size"])
df["price"] = df.apply(lambda x: qm.equities.historical.get_price(
s3_client, s3_bucket, x["symbol"], trading_day),
axis=1)
df["amount"] = (df["position_size"] / df["price"]).apply(
np.floor).astype(int)
df["txn_dollars"] = df["price"] * df["amount"] * -1
df["date"] = pd.to_datetime(arrow.get(trading_day.date()).datetime)
df.set_index("date", inplace=True)
del df["position_size"]
return df
