IndexError:索引666超出轴1的大小为501的范围

时间:2017-10-28 15:50:50

标签: python algorithm numpy

import math
import numpy as np

def ExplicitMethod(S0, K, r, q, T, Sigma, M, N, Option):

    M = int(M)
    N = int(N)
    dt = T / N
    K = float(K)
    Smax = 2 * K
    dS = Smax / N

    FGrid = np.zeros(shape=(N+1, M+1))

    if Option == 'Call':
        FGrid[-1, :] = np.maximum(np.arange(0, M+1) * dS - K, 0)  
    elif Option == 'Put':
        FGrid[-1, :] = np.maximum(K - np.arange(0, M+1) * dS, 0)

    A = np.zeros(shape=(M+1, M+1))
    A[0,0] = 1
    A[-1,-1] = 1
    for j in range(1, M-1):
        A[j, (j-1,j, j+1)] = \
        [0.5 * dt * (Sigma ** 2 * j ** 2 - (r - q) * j  ), \
         1 - dt * (Sigma ** 2 * j ** 2 + r), \
         0.5 * dt * (Sigma **2 * j ** 2 + (r - q) * j)]


    for i in range(N):
        Fhat = FGrid [i, ]
        FGrid[i, :] = np.dot(A , Fhat)
        Fhat[0] = 0
        Fhat[-1] = Smax - K * math.exp(-r * (N - i) * dt)

    k = math.floor(S0 / dS)

    V = FGrid[0, k] + (FGrid[0, k] - FGrid[0, k]) / dS * [(S0 - k * dS)]

    print (V)

ExplicitMethod (50, 30, 0.1, 0.05, 2, 0.3, 400, 800, Option = 'Call')

我收到了这个错误:

  

IndexError Traceback(最近一次调用   最后)in()        37打印(V)        38   ---> 39明确方法(50,30,0.1,0.05,2,3.3,400,800,Option ='Call')

     在ExplicitMethod中的

(S0,K,r,q,T,   Sigma,M,N,Option)        33 k = math.floor(S0 / dS)        34   ---> 35 V = FGrid [0,k] +(FGrid [0,k] - FGrid [0,k])/ dS * [(S0-k * dS)]        36        37 print(V)

     

IndexError:索引666超出了轴1的大小为401

的范围

我正在编制期权定价显式方法:

enter image description here

请帮助,谢谢!

1 个答案:

答案 0 :(得分:2)

您已将FGrid初始化为(801,401)。由于某种原因,我们告诉我们k=666

k = math.floor(S0 / dS)
V = FGrid[0, k] + (FGrid[0, k] - FGrid[0, k]) / dS * [(S0 - k * dS)]

您需要优化k的设置方式。也许数学错了。至少你需要确保它不会超过400。

In [200]: S0=50; K=30; N=800
In [201]: dS = 2*K/N
In [202]: dS
Out[202]: 0.075
In [203]: S0/dS
Out[203]: 666.6666666666667

你的666