import json from abc import ABC, abstractmethod from sys import exception import pandas as pd from dateutil.relativedelta import relativedelta from empyrical import sortino_ratio from py_jftech import filter_weekend, dict_remove, get_config, component, autowired, next_workday, \ is_workday from api import AssetOptimize, Navs, Datum, AssetPoolType, DatumType from asset_pool.dao import robo_assets_pool as rop class SortinoAssetOptimize(AssetOptimize, ABC): def __init__(self): optimize_config = get_config(__name__) self._config = [{ **x, 'name': [f"sortino_{y[1]}_{y[0]}" for y in x.items() if y[0] != 'weight'][0] } for x in optimize_config['sortino-weight']] if 'sortino-weight' in optimize_config else [] @property def delta_kwargs(self): result = [] for item in self._config: delta_kwargs = item.copy() del delta_kwargs['weight'], delta_kwargs['name'] result.append(delta_kwargs) return result def find_optimize(self, fund_ids, day): pass def get_optimize_pool(self, day): pass @property @abstractmethod def nav_min_dates(self) -> dict: pass @abstractmethod def get_groups(self): ''' :return: 返回待处理的id数组 ''' pass @abstractmethod def get_pct_change(self, fund_ids, day): ''' 根据id数组,返回指定日期的收益率 :param fund_ids: id数组 :param day: 指定的日期 :return: 收益率 ''' pass @abstractmethod def has_change(self, day): return False @component(bean_name='dividend') class FundDividendSortinoAssetOptimize(SortinoAssetOptimize): """ 根据索提诺比率计算基金优选的优选实现 以美国资产为主:US_STOCK、US_HY_BOND、US_IG_BOND Sortino ratio对资产进行排序,选出排名靠前的资产(非一类选一只) """ @autowired def __init__(self, navs: Navs = None, datum: Datum = None): super().__init__() self._navs = navs self._datum = datum self._conf = get_config(__name__) @property def asset_include(self): return self._conf['asset-include'] @property def asset_filter(self): return self._conf.get('asset-filter') @property def optimize_count(self): return self._conf['optimize-count'] @property def nav_min_dates(self) -> dict: return self._navs.get_nav_start_date() def has_change(self, day): return self._datum.update_change(day) def find_optimize(self, fund_ids, day): assert self._config, "find optimize, but not found sortino config." pct_change = pd.DataFrame(self.get_pct_change(fund_ids, day)) pct_change.set_index('date', inplace=True) sortino = pd.DataFrame() for item in self._config: ratio = dict(sortino_ratio( pct_change.truncate(before=(day - relativedelta(**dict_remove(item, ('weight', 'name'))))))) sortino = pd.concat([sortino, pd.DataFrame([ratio], index=[item['name']])]) sortino = sortino.T sortino['score'] = sortino.apply(lambda r: sum([x['weight'] * r[x['name']] for x in self._config]), axis=1) sortino.sort_values('score', ascending=False, inplace=True) # 取得分数高的前optimize_count个 return pct_change.columns[sortino.index[0:self.optimize_count]].values def get_optimize_pool(self, day): opt_pool = rop.get_one(day=day, type=AssetPoolType.OPTIMIZE) if opt_pool is not None: return json.loads(opt_pool['asset_ids']) last_one = rop.get_last_one(day=day, type=AssetPoolType.OPTIMIZE) if not last_one or day > last_one['date']: pool = [] min_dates = self.nav_min_dates max_incept_date = sorted([(day - relativedelta(**x)) for x in self.delta_kwargs])[0] max_incept_date = max_incept_date if is_workday(max_incept_date) else next_workday(max_incept_date) for fund_group in self.get_groups(): fund_group = [x for x in fund_group if min_dates[x] <= max_incept_date] if len(fund_group) > self.optimize_count: pool.extend(self.find_optimize(tuple(fund_group), day)) elif len(fund_group) <= self.optimize_count: pool.extend(fund_group) rop.insert(day, AssetPoolType.OPTIMIZE, sorted(pool)) last_one = rop.get_last_one(day=day, type=AssetPoolType.OPTIMIZE) return json.loads(last_one['asset_ids']) def get_filtered_funds(self): funds = self._datum.get_datums(type=DatumType.FUND) if self.asset_filter: filters = list(self.asset_filter.keys())[0] funds_in = [] for fund in funds: if fund[filters] in self.asset_filter[filters]: funds_in.append(fund) return funds_in return funds def get_groups(self): funds = pd.DataFrame(self.get_filtered_funds()) result = [] include = list(self.asset_include.keys())[0] for key, fund_group in funds.groupby(by=include): if key in self.asset_include[include]: result.append(tuple(fund_group['id'])) return result def get_pct_change(self, fund_ids, day): if not self._config: raise exception(f"find optimize, but not found sortino config.") start = filter_weekend( sorted([day - relativedelta(days=1, **dict_remove(x, ('weight', 'name'))) for x in self._config])[0]) fund_navs = pd.DataFrame(self._navs.get_fund_navs(fund_ids=tuple(fund_ids), min_date=start, max_date=day)) if not fund_navs.empty: fund_navs.sort_values('nav_date', inplace=True) fund_navs = fund_navs.pivot_table(index='nav_date', columns='fund_id', values='nav_cal') fund_navs.fillna(method='ffill', inplace=True) result = round(fund_navs.pct_change().dropna(), 4) result.reset_index(inplace=True) result.rename(columns={'nav_date': 'date'}, inplace=True) return result.to_dict('records') return []