MeanTesting - Code Metrics - Inspection of "mean null hypothesis testing" - chen0040/pysie - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Push — master ( f3bcee...f964b6 )

by Xianshun

created 2017-06-04 06:33 UTC

MeanTesting A

↳ Parent: Project

Complexity

Total Complexity

Size/Duplication

Total Lines	41
Duplicated Lines	68.29 %

Importance

Changes	1
Bugs	0	Features	0

Metric	Value
c	1
b	0
f	0
dl	28
loc	41
rs	10
wmc	7

2 Methods

Rating	Name	Duplication	Size	Complexity
B	__init__()	28	28	6
A	will_reject()	0	3	1

How to fix Duplicated Code

import math

from pysie.stats.distributions import DistributionFamily
from scipy.stats import norm, t


class MeanTesting(object):
    sampling_distribution = None
    p_value_one_tail = None
    p_value_two_tail = None
    mean_null = None
    test_statistic = None
    significance_level = None
    reject_mean_null = None

    def __init__(self, sampling_distribution, mean_null, significance_level=None):

        self.sampling_distribution = sampling_distribution
        self.mean_null = mean_null
        if significance_level is not None:
            self.significance_level = significance_level

        if self.sampling_distribution.distribution_family == DistributionFamily.normal:
            standard_error_null = sampling_distribution.standard_error
            Z = (sampling_distribution.point_estimate - mean_null) / standard_error_null
            self.test_statistic = Z
            pf = norm.cdf(Z)
            if Z < 0:
                pf -= 0.5
            self.p_value_one_tail = 1 - pf
            self.p_value_two_tail = self.p_value_one_tail * 2
        else:
            standard_error_null = sampling_distribution.standard_error
            td_df = (sampling_distribution.point_estimate - mean_null) / standard_error_null
            self.test_statistic = td_df
            pf = t.cdf(td_df, sampling_distribution.df)
            if td_df < 0:
                pf -= 0.5
            self.p_value_one_tail = 1 - pf
            self.p_value_two_tail = self.p_value_one_tail * 2

        if significance_level is not None:
            self.reject_mean_null = (self.p_value_one_tail < significance_level,
                                     self.p_value_two_tail < significance_level)

    def will_reject(self, significance_level):

        return self.p_value_one_tail < significance_level, self.p_value_two_tail < significance_level


class ProportionTesting(object):
    sampling_distribution = None
    p_value_one_tail = None
    p_value_two_tail = None
    mean_null = None
    test_statistic = None
    significance_level = None
    reject_mean_null = None

    def __init__(self, sampling_distribution, p_null, significance_level=None):

        self.sampling_distribution = sampling_distribution
        self.p_null = p_null
        if significance_level is not None:
            self.significance_level = significance_level

        if self.sampling_distribution.distribution_family == DistributionFamily.normal:
            standard_error_null = math.sqrt(p_null * (1 - p_null) / sampling_distribution.sample_size)
            Z = (sampling_distribution.point_estimate - p_null) / standard_error_null
            self.test_statistic = Z
            pf = norm.cdf(Z)
            if Z < 0:
                pf -= 0.5
            self.p_value_one_tail = 1 - pf
            self.p_value_two_tail = self.p_value_one_tail * 2
        else:
            standard_error_null = math.sqrt(p_null * (1 - p_null) / sampling_distribution.sample_size)
            td_df = (sampling_distribution.point_estimate - p_null) / standard_error_null
            self.test_statistic = td_df
            pf = t.cdf(td_df, sampling_distribution.df)
            if td_df < 0:
                pf -= 0.5
            self.p_value_one_tail = 1 - pf
            self.p_value_two_tail = self.p_value_one_tail * 2

        if significance_level is not None:
            self.reject_mean_null = (self.p_value_one_tail < significance_level,
                                     self.p_value_two_tail < significance_level)



1		import math
2
3		from pysie.stats.distributions import DistributionFamily
4		from scipy.stats import norm, t
5
6
7		class MeanTesting(object):
8		sampling_distribution = None
9		p_value_one_tail = None
10		p_value_two_tail = None
11		mean_null = None
12		test_statistic = None
13		significance_level = None
14		reject_mean_null = None
15
16	View Code Duplication	def __init__(self, sampling_distribution, mean_null, significance_level=None):
		0 ignored issues – show Duplication introduced 2017-06-04 06:34 UTC by Report Bug Copy Issue Report This code seems to be duplicated in your project. Loading history...
17		self.sampling_distribution = sampling_distribution
18		self.mean_null = mean_null
19		if significance_level is not None:
20		self.significance_level = significance_level
21
22		if self.sampling_distribution.distribution_family == DistributionFamily.normal:
23		standard_error_null = sampling_distribution.standard_error
24		Z = (sampling_distribution.point_estimate - mean_null) / standard_error_null
25		self.test_statistic = Z
26		pf = norm.cdf(Z)
27		if Z < 0:
28		pf -= 0.5
29		self.p_value_one_tail = 1 - pf
30		self.p_value_two_tail = self.p_value_one_tail * 2
31		else:
32		standard_error_null = sampling_distribution.standard_error
33		td_df = (sampling_distribution.point_estimate - mean_null) / standard_error_null
34		self.test_statistic = td_df
35		pf = t.cdf(td_df, sampling_distribution.df)
36		if td_df < 0:
37		pf -= 0.5
38		self.p_value_one_tail = 1 - pf
39		self.p_value_two_tail = self.p_value_one_tail * 2
40
41		if significance_level is not None:
42		self.reject_mean_null = (self.p_value_one_tail < significance_level,
43		self.p_value_two_tail < significance_level)
44
45		def will_reject(self, significance_level):
46
47		return self.p_value_one_tail < significance_level, self.p_value_two_tail < significance_level
48
49
50		class ProportionTesting(object):
51		sampling_distribution = None
52		p_value_one_tail = None
53		p_value_two_tail = None
54		mean_null = None
55		test_statistic = None
56		significance_level = None
57		reject_mean_null = None
58
59	View Code Duplication	def __init__(self, sampling_distribution, p_null, significance_level=None):
		0 ignored issues – show Duplication introduced 2017-06-04 06:34 UTC by Report Bug Copy Issue Report This code seems to be duplicated in your project. Loading history...
60		self.sampling_distribution = sampling_distribution
61		self.p_null = p_null
62		if significance_level is not None:
63		self.significance_level = significance_level
64
65		if self.sampling_distribution.distribution_family == DistributionFamily.normal:
66		standard_error_null = math.sqrt(p_null * (1 - p_null) / sampling_distribution.sample_size)
67		Z = (sampling_distribution.point_estimate - p_null) / standard_error_null
68		self.test_statistic = Z
69		pf = norm.cdf(Z)
70		if Z < 0:
71		pf -= 0.5
72		self.p_value_one_tail = 1 - pf
73		self.p_value_two_tail = self.p_value_one_tail * 2
74		else:
75		standard_error_null = math.sqrt(p_null * (1 - p_null) / sampling_distribution.sample_size)
76		td_df = (sampling_distribution.point_estimate - p_null) / standard_error_null
77		self.test_statistic = td_df
78		pf = t.cdf(td_df, sampling_distribution.df)
79		if td_df < 0:
80		pf -= 0.5
81		self.p_value_one_tail = 1 - pf
82		self.p_value_two_tail = self.p_value_one_tail * 2
83
84		if significance_level is not None:
85		self.reject_mean_null = (self.p_value_one_tail < significance_level,
86		self.p_value_two_tail < significance_level)
87
88

chen0040 / pysie

Push — master ( f3bcee...f964b6 )

MeanTesting A

Complexity

Size/Duplication

Importance

2 Methods

How to fix Duplicated Code

Duplicated Code

Duplication Side-by-Side

Filter issues like