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
|
close all;
clear; clc;
% Get all results files
files = dir("../../../results/*.out");
tw = load("../tracy-widom-approx/tracy_widom_normal.txt");
x = tw(:,1);
tw1 = tw(:,2);
nrm = tw(:,3);
N_list = []
%% Make a histogram for every results file
for f = 1:length(files)
% Get filename
filename = files(f).name;
% Extract N and L
tokens = regexp(filename,'eigs_(\d+)_(\d+)\.out', 'tokens');
N = str2double(tokens{1}{1});
N_list = [N_list, N];
L = str2double(tokens{1}{2});
% Import actual data
data = load(strcat("../../../results/",filename));
% And clip to just the eigenvalues
data = data(:,2);
% Remove outliers
data_clean = data(data > sqrt(N));
% Scale and such
s_data = (data_clean - 2*sqrt(N))*N^(1/6);
% Plot them
figure(f);
title([num2str(N)," matrix size, ",num2str(L)," data points"])
hold on
[counts, bins] = hist(s_data(s_data > -6), linspace(-6,4,40));
bin_width = bins(2)-bins(1);
pdf = counts/(sum(counts) * bin_width);
bar(bins, pdf);
plot(x,tw1);
A = [bins', pdf']
save("-ascii",["N",num2str(N),"_L",num2str(L),".txt"],"A")
end
%% Plot largest sample size for each matrix size N
skews = [];
kurts = [];
i = 0;
% Plot totals
for n = unique(N_list)
i = i + 1;
% Plot all histogram values for this N
n_files = dir(strcat("../../../results/eigs_",num2str(n),"_*.out"));
n_vals = [];
SSE = [];
% Get cumulative values
for f = 1:length(n_files)
filename = n_files(f).name;
data = load(strcat("../../../results/",filename));
data = data(:,2);
data_clean = data(data > sqrt(n));
s_data = (data_clean -2*sqrt(n)) * n^(1/6);
n_vals = [n_vals, s_data'];
end
% Plot a histogram showing the Gaussian, TW1, and sampled distribution
figure(n)
title(num2str(length(n_vals)))
hold on
% Generate the histogram
[counts, bins] = hist(n_vals(n_vals > -8), linspace(-8,6,70));
bin_width = bins(2)-bins(1);
pdf = counts/(sum(counts) * bin_width);
% And plot it
bar(bins,pdf)
plot(bins, pdf);
plot(x,nrm,'--')
plot(x,tw1);
% Save the histogram to file
A = [bins', pdf']
save("-ascii",[num2str(n),"_max.txt"], "A")
% Calculate some statistics
[skews(i),kurts(i)]=pdf_stats(bins,pdf);
end
figure(n+1)
plot(unique(N_list),skews)
hold on
plot(unique(N_list),kurts)
set(gca, "xtick", unique(N_list))
A = [unique(N_list)', skews', kurts']
save -ascii props.txt A
|
