MA342Project4/ReturnPrediction.m at main · rhit-silkaijr/MA342Project4 · 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
clear variables;
close all;
clc;

%% Historical Data
% Get all closing stock prices
[dateRange, stockIndex, P] = DataProcessing(5);
Pinit = P;

% Model parameters
numDays = 21;
numMonthsSimulated = 4;
numSimsPerMonth = 20;
numSims = 100;
investments = 0;

options = optimoptions('quadprog', 'Display', 'off');

n = 20; % number of stocks
for k=1:numSims
    P = Pinit;
    % Set initial arbitrary alpha
    alpha = 0.5;
    %alpha = rand;

    % Set initial investment value
    investment = 10000;
    expReturn = zeros(numMonthsSimulated,numSimsPerMonth); % all expected simulated returns
    for i=1:numMonthsSimulated

        % Historical return ratio between days
        R = (P(2:end,:)-P(1:end-1,:))./P(1:end-1,:);

        % average return over entire period for all stocks
        r = mean(R);

        % the covariance of all our return ratios
        C = cov(R);

        if i ~= 1
            wOld = w;
        end

        % Unlimited optimization
        % [w, optVal] = quadprog((1-alpha)*2*C, -alpha*r, [], [], ones(1,n), [1], zeros(n,1), ones(n,1), [], options);

        % Limited optimization
        [w, optVal] = quadprog((1-alpha)*2*C, -alpha*r, [], [], ones(1,n), [1], zeros(n,1), 0.2*ones(n,1), [], options);

        % 5% cost on trades
        if i ~= 1
            deltaW = w - wOld;
            investment = investment - sum(abs(deltaW) * 0.05 * investment);
        end

        % Repeatedly simulate future stock prices and assess performance
        for j=1:numSimsPerMonth

            % generate future prices
            Po = StockSimulation(P,numDays);

            % Historical return ratio between days
            R = (Po(2:end,:)-Po(1:end-1,:))./Po(1:end-1,:);

            % average return over entire period for all stocks
            r = mean(R);

            expReturn(i,j) = r * w;

        end

        % "Real" next month stock prices
        Po = StockSimulation(P,numDays);
        P = [P(numDays+1:length(P),:)' flip(Po(2:numDays+1,:))' ]';

        % Actual return based on simulated price

        R = (Po(2:end,:)-Po(1:end-1,:))./Po(1:end-1,:);

        r = mean(R);

        %Assess performance based on multiple simulations
        simReturn = mean(expReturn(1,i));

        % Options:
        % Don't change alpha
        % alpha = alpha;
        % Increase alpha if profit, decrease if loss
        if simReturn > 0
            alpha = min(1, alpha + 0.1);
        elseif simReturn < 0
            alpha = max(0, alpha - 0.1);
        end
        % Randomize alpha every month
        % alpha = rand;

        % Update investment

        actualReturn = r * w;
        investment = investment * (1 + actualReturn);
    end

    %figure
    %plot(P)
    investments = investments + investment;
    %fprintf("After %d months, $10000 has been transformed into $%f\n", numMonthsSimulated, investment);
end
investments = investments / numSims;
fprintf("With %d months and %d runs, \n$10000 is on average transformed into $%f\n", numMonthsSimulated, numSims, mean(investments));