-
Notifications
You must be signed in to change notification settings - Fork 20
/
ResultTables.m
148 lines (140 loc) · 9.31 KB
/
ResultTables.m
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
136
137
138
139
140
141
142
143
144
145
146
147
148
function ResultTables(param,betavec,result_js,result_jr87,dopltos)
%this function makes nice tables of results from the MC run
%it can be called during the MC loop to save intermediate results on disk
%number of produced results for different betas
numbetas=min([numel(betavec);numel(result_js);numel(result_jr87)]);
fprintf('\n\n\n');
diary 'GrandResultsMC.txt'
disp(datestr(now));
fprintf('Table 1: Estimation results (RMSE in brackets)\n');
fprintf('----------------------------------------------------------------------------------------------------------------------------------------------------------\n');
fprintf('%20s %18s %18s %18s %18s %18s %18s %18s\n','Method','RC','c','p1','p2','p3','p4','p5');
fprintf('----------------------------------------------------------------------------------------------------------------------------------------------------------\n');
fprintf('%20s %18.4f %18.4f %18.4f %18.4f %18.4f %18.4f %18.4f\n','True values --> ',param.RC,param.thetaCost,param.thetaProbs);
fprintf('----------------------------------------------------------------------------------------------------------------------------------------------------------\n');
for i=1:numbetas
fprintf('beta = %1.5f\n',betavec(i));
%corrections of the results for display
%add zeros if dimentions are too low
if sum(size(result_js{i}.mean_estimates)<[7 4])
result_js{i}.mean_estimates(7,4)=0;
end
if sum(size(result_js{i}.std_estimates)<[7 3])
result_js{i}.std_estimates(7,3)=0;
end
if size(result_jr87{i}.p.est,2)<5
result_jr87{i}.p.est(size(result_jr87{i}.p.est,1),5)=0;
end
fprintf('----------------------------------------------------------------------------------------------------------------------------------------------------------\n');
js_method=1;%MPEC/AMPL
fprintf('%20s %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f)\n',...
'MPEC/AMPL',...
result_js{i}.mean_estimates(7,js_method+1),result_js{i}.rmse_estimates(7,js_method),...
result_js{i}.mean_estimates(1,js_method+1),result_js{i}.rmse_estimates(1,js_method),...
result_js{i}.mean_estimates(2,js_method+1),result_js{i}.rmse_estimates(2,js_method),...
result_js{i}.mean_estimates(3,js_method+1),result_js{i}.rmse_estimates(3,js_method),...
result_js{i}.mean_estimates(4,js_method+1),result_js{i}.rmse_estimates(4,js_method),...
result_js{i}.mean_estimates(5,js_method+1),result_js{i}.rmse_estimates(5,js_method),...
result_js{i}.mean_estimates(6,js_method+1),result_js{i}.rmse_estimates(6,js_method)...
)
js_method=2;%MPEC/ktrlink
fprintf('%20s %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f)\n',...
'MPEC/ktrlink',...
result_js{i}.mean_estimates(7,js_method+1),result_js{i}.rmse_estimates(7,js_method),...
result_js{i}.mean_estimates(1,js_method+1),result_js{i}.rmse_estimates(1,js_method),...
result_js{i}.mean_estimates(2,js_method+1),result_js{i}.rmse_estimates(2,js_method),...
result_js{i}.mean_estimates(3,js_method+1),result_js{i}.rmse_estimates(3,js_method),...
result_js{i}.mean_estimates(4,js_method+1),result_js{i}.rmse_estimates(4,js_method),...
result_js{i}.mean_estimates(5,js_method+1),result_js{i}.rmse_estimates(5,js_method),...
result_js{i}.mean_estimates(6,js_method+1),result_js{i}.rmse_estimates(6,js_method)...
)
js_method=3;%NFXP-SA with ktrlink
fprintf('%20s %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f)\n',...
'NFXP-SA/ktrlink',...
result_js{i}.mean_estimates(7,js_method+1),result_js{i}.rmse_estimates(7,js_method),...
result_js{i}.mean_estimates(1,js_method+1),result_js{i}.rmse_estimates(1,js_method),...
result_js{i}.mean_estimates(2,js_method+1),result_js{i}.rmse_estimates(2,js_method),...
result_js{i}.mean_estimates(3,js_method+1),result_js{i}.rmse_estimates(3,js_method),...
result_js{i}.mean_estimates(4,js_method+1),result_js{i}.rmse_estimates(4,js_method),...
result_js{i}.mean_estimates(5,js_method+1),result_js{i}.rmse_estimates(5,js_method),...
result_js{i}.mean_estimates(6,js_method+1),result_js{i}.rmse_estimates(6,js_method)...
)
fprintf('%20s %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f) %8.4f(%8.4f)\n',...
'NFXP-NK/BHHH',...
mean(result_jr87{i}.RC.est),sqrt(mean((result_jr87{i}.RC.est-param.RC).^2)),...
mean(result_jr87{i}.c.est),sqrt(mean((result_jr87{i}.c.est-param.thetaCost).^2)),...
mean(result_jr87{i}.p.est(:,1)),sqrt(mean((result_jr87{i}.p.est(:,1)-param.thetaProbs(1)).^2)),...
mean(result_jr87{i}.p.est(:,2)),sqrt(mean((result_jr87{i}.p.est(:,2)-param.thetaProbs(2)).^2)),...
mean(result_jr87{i}.p.est(:,3)),sqrt(mean((result_jr87{i}.p.est(:,3)-param.thetaProbs(3)).^2)),...
mean(result_jr87{i}.p.est(:,4)),sqrt(mean((result_jr87{i}.p.est(:,4)-param.thetaProbs(4)).^2)),...
mean(result_jr87{i}.p.est(:,5)),sqrt(mean((result_jr87{i}.p.est(:,5)-param.thetaProbs(5)).^2))...
)
fprintf('----------------------------------------------------------------------------------------------------------------------------------------------------------\n');
end
fprintf('\n\n');
fprintf('Table 2: Numerical performance\n');
fprintf('-----------------------------------------------------------------------------------------------------------------------\n');
fprintf('%14s %14s %14s %14s %14s %14s %14s\n', '' , 'Runs Converged', 'CPU Time' ,'# of Major' ,'# of Func.' ,'# of Contract','# of N-K');
fprintf('%-14s %14s %14s %14s %14s %14s %14s\n', 'beta', sprintf('(out of %g)',param.MC*param.multistarts) ,'(in sec.)','Iter' ,'Eval.' ,'Iter.','Iter.')
fprintf('-----------------------------------------------------------------------------------------------------------------------\n');
fprintf(' MPEC/AMPL\n');
fprintf('-----------------------------------------------------------------------------------------------------------------------\n');
for i=1:numbetas
fprintf('%-14.4g %14d %14.4f %14.4f %14.4f %14s %14s\n', ...
betavec(i), result_js{i}.TotalSuccess(1) ,mean(result_js{i}.runtime(result_js{i}.converged(:,1)==1,1)) , ...
result_js{i}.num_iter(1),result_js{i}.ave_feval(1), '-','-')
end
fprintf('-----------------------------------------------------------------------------------------------------------------------\n');
fprintf(' MPEC/ktrlink\n');
fprintf('-----------------------------------------------------------------------------------------------------------------------\n');
for i=1:numbetas
fprintf('%-14.4g %14d %14.4f %14.4f %14.4f %14s %14s\n', ...
betavec(i), result_js{i}.TotalSuccess(2) ,mean(result_js{i}.runtime(result_js{i}.converged(:,2)==1,2)) , ...
result_js{i}.num_iter(2),result_js{i}.ave_feval(2), '-','-')
end
fprintf('-----------------------------------------------------------------------------------------------------------------------\n');
fprintf(' NFXP-SA with ktrlink\n');
fprintf('-----------------------------------------------------------------------------------------------------------------------\n');
for i=1:numbetas
fprintf('%-14.4g %14d %14.4f %14.4f %14.4f %14.4f %14s\n', ...
betavec(i), result_js{i}.TotalSuccess(3) ,mean(result_js{i}.runtime(result_js{i}.converged(:,3)==1,3)) , ...
result_js{i}.num_iter(3),result_js{i}.ave_feval(3), result_js{i}.ave_bellmaniter,'-')
end
fprintf('-----------------------------------------------------------------------------------------------------------------------\n');
fprintf(' NFXP-NK with 2 step ML and BHHH\n');
fprintf('-----------------------------------------------------------------------------------------------------------------------\n');
for i=1:numbetas
fprintf('%-14.4g %14d %14.4f %14.4f %14.4f %14.4f %14.4f\n', ...
betavec(i), result_jr87{i}.TotalSuccess ,mean(result_jr87{i}.runtime(result_jr87{i}.converged)) , ...
mean(result_jr87{i}.MajorIter),mean(result_jr87{i}.FuncEval), ...
mean(result_jr87{i}.BellmanIter),mean(result_jr87{i}.NKIter))
end
fprintf('-----------------------------------------------------------------------------------------------------------------------\n');
fprintf('\n');
fprintf('Remaining parameters\n');
fprintf('RC = %10.5f \n',param.RC);
fprintf('c = %10.5f \n',param.thetaCost);
for i= 1:numel(param.thetaProbs)
fprintf('p(%d) = %10.5f \n',i, param.thetaProbs(i));
end
fprintf('n = %10.5f \n',param.N);
fprintf('N = %10.5f \n',param.nBus);
fprintf('T = %10.5f \n',param.nT);
fprintf('\n\n');
diary off
if dopltos
for i=1:numbetas
fig=figure('Color',[1 1 1],'NextPlot','new');
ax=axes('Parent',fig);
xx=(0.001:0.001:1)';
data{1}=result_js{i}.runtime(result_js{i}.converged(:,1)==1,1);
hold on
plot(quantile(data{1},xx),xx, '-k','DisplayName','MPEC/AMPL','Parent',ax);
plot(quantile(result_jr87{i}.runtime,xx),xx, '-r','DisplayName','NFXP-NK with 2 stage ML and BHHH','Parent',ax);
xlabel('CPU time (seconds)')
legend1 = legend(ax,'show');
set(legend1,'EdgeColor',[1 1 1],'Location','SouthEast','YColor',[1 1 1],'XColor',[1 1 1]);
title(sprintf('Distribution of estimation CPU Time with beta=%1.5f (conditional on converging)',betavec(i)));
end
end
end