Coverage for src/cvxmarkowitz/models/trading_costs.py: 100%

1# Copyright 2023 Stanford University Convex Optimization Group 

2# 

3# Licensed under the Apache License, Version 2.0 (the "License"); 

4# you may not use this file except in compliance with the License. 

5# You may obtain a copy of the License at 

6# 

7# http://www.apache.org/licenses/LICENSE-2.0 

8# 

9# Unless required by applicable law or agreed to in writing, software 

10# distributed under the License is distributed on an "AS IS" BASIS, 

11# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 

12# See the License for the specific language governing permissions and 

13# limitations under the License. 

14"""Model for trading costs.""" 

15 

16from __future__ import annotations 

17 

18from dataclasses import dataclass 

19 

20import cvxpy as cp 

21import numpy as np 

22 

23from cvxmarkowitz.model import Model 

24from cvxmarkowitz.names import DataNames as D 

25from cvxmarkowitz.names import ParameterName as P 

26from cvxmarkowitz.types import Dimensions, Matrix, Variables 

27from cvxmarkowitz.utils.fill import fill_vector 

28 

29 

30@dataclass(frozen=True) 

31class TradingCosts(Model): 

32 """Model for trading costs.""" 

33 

34 def __post_init__(self) -> None: 

35 """Initialize trading cost parameters and previous-weights cache.""" 

36 self.parameter[P.POWER] = cp.Parameter(shape=(), name=P.POWER, value=1.0) 

37 

38 # initial weights before rebalancing -- keyed by D.WEIGHTS, the same name 

39 # the decision variable uses, since it is the previous value of it. 

40 self.data[D.WEIGHTS] = cp.Parameter(shape=self.assets, name=D.WEIGHTS, value=np.zeros(self.assets)) 

41 

42 def estimate(self, variables: Variables) -> cp.Expression: 

43 """Estimate trading costs for a rebalance. 

44 

45 Args: 

46 variables: Optimization variables, expected to contain D.WEIGHTS. 

47 

48 Returns: 

49 A convex expression representing the p-power cost of trades 

50 between current and previous weights. 

51 """ 

52 return cp.sum( 

53 cp.power( 

54 cp.abs(variables[D.WEIGHTS] - self.data[D.WEIGHTS]), 

55 p=self.parameter[P.POWER], 

56 ) 

57 ) 

58 

59 def dimensions(self, **kwargs: Matrix) -> Dimensions: 

60 """Return the number of assets the previous weights imply.""" 

61 return ((D.WEIGHTS, len(kwargs[D.WEIGHTS])),) 

62 

63 def update(self, **kwargs: Matrix) -> None: 

64 """Update cached data values. 

65 

66 Expected keyword arguments: 

67 weights: Vector of previous weights used as the trading baseline. 

68 """ 

69 self.data[D.WEIGHTS].value = fill_vector(num=self.assets, x=kwargs[D.WEIGHTS])