/
runMultinestStats.py
195 lines (141 loc) · 6.15 KB
/
runMultinestStats.py
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
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
from numpy import *
from scipy.special import erfinv, gammaln
from scipy.stats import skewnorm, norm, truncnorm
import matplotlib.pyplot as plt
import pymultinest
from pymultinest import Analyzer
import mpi4py
import json
import time
# Read in data
bkgpdf_data = genfromtxt("Data/bkgpdf.txt")
brnpdf_data = genfromtxt("Data/brnpdf.txt")
brndelayedpdf_data = genfromtxt("Data/delbrnpdf.txt")
cevnspdf_data = genfromtxt("Data/cevnspdf.txt")
obs_data = genfromtxt("Data/datanobkgsub.txt")
# Set up CEvNS, BRN, and Observed arrays
brn_prompt = brnpdf_data[:,3]
brn_delayed = brndelayedpdf_data[:,3]
obs = obs_data[:,3]
cevns = cevnspdf_data[:,3]
ss = bkgpdf_data[:,3]
# Flat bins
entries = obs_data.shape[0]
keVee = obs_data[:,0]
f90 = obs_data[:,1]
timing = obs_data[:,2]
# Define stats CDFs for priors
ss_error = sqrt(sum(ss)/5)/sum(ss) # percent error
normSS = norm(scale=ss_error) #truncnorm(-1.0,1.0,scale=ss_error)
normPromptBRN = norm(scale=0.3) #truncnorm(-1.0,1.0,scale=0.3)
normDelayedBRN = norm(scale=1.0) #truncnorm(-1.0,1.0,scale=1.0)
# Define Priors for MultiNest
def prior_stat(cube, n, d):
cube[0] = 2*cube[0] # CEvNS norm
cube[1] = normSS.ppf(cube[1]) # SS norm
cube[2] = normPromptBRN.ppf(cube[2]) # BRN prompt norm
cube[3] = normDelayedBRN.ppf(cube[3]) # BRN delayed norm
def prior_stat_null(cube, n, d):
cube[0] = normSS.ppf(cube[0]) # SS norm
cube[1] = normPromptBRN.ppf(cube[1]) # BRN prompt norm
cube[2] = normDelayedBRN.ppf(cube[2]) # BRN delayed norm
# Generate new PDFs with nuisance-controlled norms
def events_gen_stat(cube, report_stats=False):
brn_prompt_syst = ((1+cube[2])*brn_prompt).clip(min=0.0001)
brn_del_syst = ((1+cube[3])*brn_delayed).clip(min=0.0001)
cevns_syst = (cube[0]*cevns).clip(min=0.0001)
ss_syst = ((1+cube[1])*ss).clip(min=0.0001)
if report_stats:
print("N_CEvNS = ", sum(cevns_syst))
print("N_BRN_PRO = ", sum(brn_prompt_syst))
print("N_BRN_DEL = ", sum(brn_del_syst))
print("N_SS = ", sum(ss_syst))
return brn_prompt_syst + brn_del_syst + cevns_syst + ss_syst
# Generate new PDFs with nuisance-controlled norms (no CEvNS)
def events_gen_stat_null(cube):
brn_prompt_syst = ((1+cube[1])*brn_prompt).clip(min=0.0001)
brn_del_syst = ((1+cube[2])*brn_delayed).clip(min=0.0001)
ss_syst = ((1+cube[0])*ss).clip(min=0.0001)
return brn_prompt_syst + brn_del_syst + ss_syst
def poisson(obs, theory):
ll = obs * log(theory) - theory - gammaln(obs+1)
return sum(ll)
def PrintSignificance():
# Print out totals.
print("TOTALS:")
print("N_obs = ", sum(obs))
print("N_ss = ", sum(ss))
print("N_brn =", sum(brn_prompt + brn_delayed))
print("N_cevns = ", sum(cevns))
an = Analyzer(4, "multinest/cenns10_stat/cenns10_stat")
bf = an.get_best_fit()['parameters']
an_null = Analyzer(3, "multinest/cenns10_stat_no_cevns/cenns10_stat_no_cevns")
bf_null = an_null.get_best_fit()['parameters']
bf = [an.get_stats()['marginals'][0]['median'], an.get_stats()['marginals'][1]['median'],
an.get_stats()['marginals'][2]['median'], an.get_stats()['marginals'][3]['median']]
bf_null = [an_null.get_stats()['marginals'][0]['median'], an_null.get_stats()['marginals'][1]['median'],
an_null.get_stats()['marginals'][2]['median']]
# Save best-fit (MLE) parameters from MultiNest (in <out>stats.dat)
# Truncated gaussian
bf_norm = [0.128203949389575733E+01,
-0.757751720547599188E-02,
0.928830540200280969E-01,
-0.681121212215910043E+00]
bf_norm_null = [-0.799580130637969101E-02,
0.253213583049654078E+00,
-0.514351228113789194E+00]
# Unconstrained Gaussian
bf_truncnorm = [0.168960153287222759E+01,
-0.312937517992761469E-01,
0.780942325684447630E-01,
-0.970385882467374672E+00]
bf_truncnorm_null = [-0.147905160635425168E-01,
0.245574237324468231E+00,
-0.460897530294036739E+00]
# Get ratio test
print("Significance (stat):")
stat_q = sqrt(abs(2*(-poisson(obs, events_gen_stat(bf)) \
+ poisson(obs, events_gen_stat_null(bf_null)))))
print(stat_q)
print("Best-fit norms:")
events_gen_stat(bf, report_stats=True)
def RunMultinest():
def loglike(cube, ndim, nparams):
n_signal = events_gen_stat(cube)
ll = obs * log(n_signal) - n_signal - gammaln(obs+1)
return sum(ll)
save_str = "cenns10_stat"
out_str = "multinest/" + save_str + "/" + save_str
json_str = "multinest/" + save_str + "/params.json"
# Run the sampler with CEvNS, BRN, and SS.
pymultinest.run(loglike, prior_stat, 4,
outputfiles_basename=out_str,
resume=False, verbose=True, n_live_points=1000, evidence_tolerance=0.5,
sampling_efficiency=0.8)
# Save the parameter names to a JSON file.
params_stat = ["cevns_norm", "ss_norm", "BRN_prompt_norm", "BRN_delayed_norm"]
json.dump(params_stat, open(json_str, 'w'))
def RunMultinestNull():
def loglike(cube, ndim, nparams):
n_signal = events_gen_stat_null(cube)
ll = obs * log(n_signal) - n_signal - gammaln(obs+1)
return sum(ll)
save_str = "cenns10_stat_no_cevns"
out_str = "multinest/" + save_str + "/" + save_str
json_str = "multinest/" + save_str + "/params.json"
# Run the sampler with just BRN, and SS.
pymultinest.run(loglike, prior_stat_null, 3,
outputfiles_basename=out_str,
resume=False, verbose=True, n_live_points=1000, evidence_tolerance=0.5,
sampling_efficiency=0.8)
# Save the parameter names to a JSON file.
params_stat_null = ["ss_norm", "BRN_prompt_norm", "BRN_delayed_norm"]
json.dump(params_stat_null, open(json_str, 'w'))
if __name__ == '__main__':
PrintSignificance()
print("Running MultiNest with CEvNS, BRN, and SS components...")
#RunMultinest()
print("Starting next run for only BRN and SS components (5s)...")
time.sleep(5.0)
#RunMultinestNull()
PrintSignificance()