1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
|
#!/usr/bin/python3
# Copyright (C) 2016 EDF
# All Rights Reserved
# This code is published under the GNU Lesser General Public License (GNU LGPL)
import numpy as np
import unittest
import random
import math
# unit test for continuation values
##################################
class testContValues(unittest.TestCase):
# test a regular grid for stocks and a local function basis for regression
def testSimpleGridsAndRegressor(self):
import StOptGrids
import StOptReg
# low value for the meshes
lowValues =np.array([1.,2.,3.],dtype=float)
# size of the meshes
step = np.array([0.7,2.3,1.9],dtype=float)
# number of steps
nbStep = np.array([3,2,4], dtype=np.int32)
# create the regular grid
#########################
grid = StOptGrids.RegularSpaceGrid(lowValues,step,nbStep)
# simulation
nbSimul =10000
np.random.seed(1000)
x = np.random.uniform(-1.,1.,size=(1,nbSimul));
# mesh
nbMesh = np.array([16],dtype=np.int32)
# Create the regressor
#####################
regressor = StOptReg.LocalLinearRegression(False,x,nbMesh)
# regressed values
toReal = (2+x[0,:]+(1+x[0,:])*(1+x[0,:]))
# function to regress
toRegress = toReal + 4*np.random.normal(0.,1,nbSimul)
# create a matrix (number of stock points by number of simulations)
toRegressMult = np.zeros(shape=(len(toRegress),grid.getNbPoints()))
for i in range(toRegressMult.shape[1]):
toRegressMult[:,i] = toRegress
# Now create the continuation object
####################################
contOb = StOptReg.ContinuationValue(grid,regressor,toRegressMult)
# get back the regressed values at the point stock
ptStock= np.array([1.2,3.1,5.9],dtype=float)
regressValues = contOb.getAllSimulations(ptStock)
# do the same with an interpolator
interp = grid.createInterpolator(ptStock)
regressValuesInterp = contOb.getAllSimulations(interp)
# test create of an interpoaltion object mixing grids for stocks and regression for uncertainties
#################################################################################################
gridAndRegressed = StOptReg.GridAndRegressedValue(grid,regressor,toRegressMult)
# get back the regressed value for a point stock and an uncertainty
valRegressed = gridAndRegressed.getValue(ptStock, x[:,0])
# Now test simulations one by one
#################################
for i in range(nbSimul):
regressed = gridAndRegressed.getValue(ptStock,x[:,i] )
diff = regressed -regressValues[i]
self.assertAlmostEqual(diff,0., 7, "test regression simulation by simulation")
# test some mapping of GneralSpaceGrid
def testGeneralGridInheritance(self):
from StOptGrids import GeneralSpaceGrid, RegularSpaceGrid
from StOptReg import LocalLinearRegression, ContinuationValue
x = np.random.randn(5)
regressor = LocalLinearRegression([1])
regular = RegularSpaceGrid(np.array([0.]), np.array([0.5]), np.array([3]))
ContinuationValue(regular, regressor, x)
general = GeneralSpaceGrid([[0., 1., 1.2, 1.5]])
ContinuationValue(general, regressor, x)
if __name__ == '__main__':
unittest.main()
|
