prediction-vector/pricing.m at master · colman01/prediction-vector · GitHub

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
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
function [ price ] = pricing( n,N,r,Szenario)
%PRICING Pricing of American Options via regression
% n is the number of simulated paths
% N is number of evaluations on new data
% r is interest rate
%  n=2000;
%  N=10000;
%  r=0.05;
%  Szenario=1;

x0=100; %starting value of the stocks

T=1;    %upper bound on time

%different number of underlying stocks
if (Szenario==1)
    m=48;   %number of time steps

    d=5;
    sigma=[0.3024 , 0.1354 , 0.0722 , 0.1367 , 0.1641;
      0.1354 , 0.2270 , 0.0613 , 0.1264 , 0.1610 ;
      0.0722 , 0.0613 , 0.0717 , 0.0884 , 0.0699 ;
      0.1367 , 0.1264 , 0.0884 , 0.2937 , 0.1394 ;
      0.1641 , 0.1610 , 0.0699 , 0.1394 , 0.2535];  %volatility matrix
else
	m=12;

    d=1;
	sigma=0.25;
end


estimates=zeros(N,m-1);
% contains already computed estimates

pathvectorPredict=paths(N,d,T,m,m,sigma,r,x0);


display('Estimate q_t')


for t=(m-1):(-1):1

	display(t)
	%Construct paths for estimation:
	pathvector_new=paths(n,d,T,m,t+1,sigma,r,x0);

	%regression:
    x_new=pathvector_new(:,:,index(t+1));
	if (Szenario==1)
        y_new=payoff1(x_new,t+1,r,m);
    else
        y_new=payoff2(x_new,t+1,r,m);
    end

	if t<(m-1)
		%q_t+1:
		Dummy=regression(x_old,y_old,[x_new;pathvectorPredict(:,:,index(t+1))]);
		y_new=max(y_new,Dummy(1:n));
		estimates(:,t+1)=Dummy((n+1):(n+N));
    end

	x_old=pathvector_new(:,:,index(t));
	y_old=y_new;

end


estimates(:,0+1)=regression(x_old,y_old,pathvectorPredict(:,:,index(0+1)));


% compute estimate for t=0:
t=0;
pathvector_new=paths(n,d,T,m,t+1,sigma,r,x0);
x_new=pathvector_new(:,:,index(t+1));
if (Szenario==1)
	y_new=payoff1(x_new,t+1,r,m);
else
	y_new=payoff2(x_new,t+1,r,m);
end

q=regression(x_old,y_old,x_new);
y_new=max(y_new,q);

estimate0=mean(y_new);


% evaluate stopping rule

display('Evaluate stopping rule')

sold=zeros(N,1);
money=zeros(N,1);

for t=0:m

	display(t)

	if ((0<t) && (t<m))
		estimate_akt=estimates(:,t);
    end

    for i=1:N

		if (sold(i)==0)

			% option not sold yet
			% compute payoff if we sell now
			if (Szenario==1)
				p=payoff1(pathvectorPredict(i,1:d,index(t)),t,r,m);
			else
				p=payoff2(pathvectorPredict(i,1:d,index(t)),t,r,m);
            end
			if (t==0)
				qt=estimate0;
            elseif (t==m)
				qt=0;
			else
				qt=estimate_akt(i);
            end

            if (qt<=p)

				sold(i)=1;
				money(i)=p;
            end

        end
    end
end

display(['Expected payoff is ' num2str(mean(money))])
price=mean(money);
end