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
|
#!/usr/bin/awk -f
### quad_reg.awk
# quadratic regression along columns
BEGIN {
OFS = ":"
pi = 4.0*atan2(1.0, 1.0)
sign = "[+-]?"
decimal = "[0-9]+[.]?[0-9]*"
fraction = "[.][0-9]*"
exponent = "([Ee]" sign "[0-9]+)?"
number = "^" sign "(" decimal "|" fraction ")" exponent "$"
}
NR == 1 {
for (n=1; n<=NF; n++)
($n ~ number) ? header[n] = "col" n : header[n] = $n
}
NF > 0 {
if (NF > nf_max)
nf_max = NF
### iterate over columns
for (y=1; y<=nf_max; y++) {
if ($y == header[n])
continue
if ($y ~ number) {
### mean
count[y] += 1
sum[y] += $y
sum2[y] += $y*$y
sum3[y] += $y*$y*$y
sum4[y] += $y*$y*$y*$y
delta0[y] = $y - mean[y]
mean[y] += delta0[y]/count[y]
mean2[y] = sum2[y]/count[y]
delta1[y] = $y - mean[y]
delta2[y] = delta0[y]*delta1[y]
sum_delta[y] += delta1[y]
sum_delta2[y] += delta2[y]
### sample variance
#(count[y] > 1) ? var[y] = sum_delta2[y]/(count[y] - 1) : var[y] = ""
# x = row, y = col, trendline: y = A + Bx + Cx^2
for (x=1; x<=nf_max; x++) {
count_xy[x,y] += 1
sum_xy[x,y] += $x*$y
sum_x2y[x,y] += $x*$x*$y
# covariances
if (count_xy[x,y] > 0) {
s_xx[x,y] = sum2[x]/count_xy[x,y] - mean[x]*mean[x]
s_xy[x,y] = sum_xy[x,y]/count_xy[x,y] - mean[x]*mean[y]
s_xx2[x,y] = sum3[x]/count_xy[x,y] - mean[x]*mean2[x]
s_x2x2[x,y] = sum4[x]/count_xy[x,y] - mean2[x]*mean2[x]
s_x2y[x,y] = sum_x2y[x,y]/count_xy[x,y] - mean2[x]*mean[y]
}
else {
s_xx[x,y] = 0
s_xy[x,y] = 0
s_xx2[x,y] = 0
s_x2x2[x,y] = 0
s_x2y[x,y] = 0
}
bc_den[x,y] = (s_xx[x,y]*s_x2x2[x,y] - s_xx2[x,y]*s_xx2[x,y])
if (bc_den[x,y] != 0) {
c[x,y] = (s_x2y[x,y]*s_xx[x,y] - s_xy[x,y]*s_xx2[x,y])/bc_den[x,y]
b[x,y] = (s_xy[x,y]*s_x2x2[x,y] - s_x2y[x,y]*s_xx2[x,y])/bc_den[x,y]
}
else {
c[x,y] = 0
b[x,y] = 0
}
a[x,y] = mean[y] - b[x,y]*mean[x] - c[x,y]*mean2[x]
# error estimate
err[x,y] = $y - a[x,y] - b[x,y]*$x - c[x,y]*$x*$x
sum_err2[x,y] += err[x,y]*err[x,y]
# correlation
sum_delta2[y] ? r2[x,y] = 1 - sum_err2[x,y]/sum_delta2[y] : r2[x,y] = 0
#print a[x,y],b[x,y],c[x,y],err[x,y],sum_err2[x,y]
# quadratic roots (x-intercepts)
if (c[x,y] != 0) {
D[x,y] = b[x,y]*b[x,y] - 4.0*a[x,y]*c[x,y]
if (D[x,y] >= 0) {
rx0[x,y] = 0.5*(-1.0*b[x,y] - sqrt(D[x,y]))/c[x,y]
rx1[x,y] = 0.5*(-1.0*b[x,y] + sqrt(D[x,y]))/c[x,y]
}
if (D[x,y] < 0) {
rx_real[x,y] = -0.5*b[x,y]/c[x,y]
rx_imag[x,y] = 0.5*sqrt(-1.0*D[x,y])/c[x,y]
rx0[x,y] = rx_real[x,y]
rx1[x,y] = rx_real[x,y]
}
# vertex of parabola
xv[x,y] = -0.5*b[x,y]/c[x,y]
yv[x,y] = -0.25*(b[x,y]*b[x,y])/c[x,y] + a[x,y]
# focus of parabola
xf[x,y] = -0.5*b[x,y]/c[x,y]
yf[x,y] = (4.0*a[x,y]*c[x,y] - b[x,y]*b[x,y] + 1.0)/(4.0*c[x,y])
}
}
}
else
continue
}
}
END {
for (x=1; x<=nf_max; x++) {
for (y=1; y<=nf_max; y++) {
if (x != y) {
printf(OFMT OFS "(%s)" OFS "= (" OFMT ")(%s)^2 + (" OFMT ")(%s) + (" OFMT ")" OFS,
r2[x,y], header[y], c[x,y], header[x], b[x,y], header[x], a[x,y])
printf("[" OFMT "," OFMT "][" OFMT "," OFMT "][" OFMT "," OFMT "]" OFS "[" OFMT "," OFMT "]",
rx0[x,y], (a[x,y] + b[x,y]*rx0[x,y] + c[x,y]*rx0[x,y]*rx0[x,y]),
rx1[x,y], (a[x,y] + b[x,y]*rx1[x,y] + c[x,y]*rx1[x,y]*rx1[x,y]),
0, a[x,y],
xv[x,y], yv[x,y])
printf(OFS "[" OFMT "," OFMT "]", xf[x,y], yf[x,y])
printf(OFS "[" OFMT "]" ORS, (yf[x,y] - yv[x,y]))
}
}
}
}
|
