roads-package: roads: An set of helper functions to take advantage of US...

Description Details Author(s) See Also Examples

Description

The R package modes was designed with a dual purpose of accurately estimating the mode (or modes) as well as characterizing the modality of data. The specific application area includes complex or mixture distibutions particularly in a big data environment. The heterogenous nature of (big) data may require deep introspective statistical and machine learning techniques, but these statistical tools often fail when applied without first understanding the data. In small datasets, this often isn't a big issue, but when dealing with large scale data analysis or big data thoroughly inspecting each dimension typically yields an O(n^n-1) problem. As such, dealing with big data require an alternative toolkit. This package not only identifies the mode or modes for various data types, it also provides a programmatic way of understanding the modality (i.e. unimodal, bimodal, etc.) of a dataset (whether it's big data or not). See <http://www.sdeevi.com/roads_package> or <http://sathish-deevi.github.io/Roads-Package> for examples and discussion.

Details

This package was designed to find the modes and aide in assessing the modality of a dataset. It was optimized programmatically to be as efficient on big data as possible. The enclosed techniques span various fields of statistics and machine learning from exploratory data analysis, to distribution theory, to univariate & multivariate statistics as well as data munging and multi-stage machine learning.

The key functions that are included in this package include:

Nonparametric

Parametric

Author(s)

Sathish Deevi & 4D Strategies

See Also

http://www.sdeevi.com/modes_package

Examples

  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
 #12 Examples of the most useful and common features of this package
#are included below.

#####		Nonparametric Examples		#####

### 1) Mode examples

 ##Example 1.1
 #data<-c(rep(6,9),rep(3,3))
 #mode(data,type=1,"NULL","NULL")

 ##Example 1.2
 #data<-c(rep(6,9),rep(3,9))
 #mode(data,type=1,"NULL","NULL")
 
 ##Example 1.3
 #data<-c(rep(6,9),rep(3,8),rep(7,7),rep(2,6))
 #mode(data,type=1,"NULL",2)

 ##Example 1.4
 #data<-c(rnorm(15,0,1),rnorm(21,5,1),rep(3,3))
 #mode(data)

 ##Example 1.5
 #data<-c(rep(6,3),rep(3,3),rnorm(15,0,1))
 #mode(data,3,NULL,4)
 #mode(data,type=2,digits=1,3)


### 2) Other General Parametric Examples

 ##Example 2.1
 #data<-c(rnorm(15,0,1),rnorm(21,5,1))
 #hist(data)
 #bimodality_amplitude(data,TRUE)
 #bimodality_coefficient(data,TRUE)
 #bimodality_ratio(data,FALSE)

 ##Example 2.2
 #data<-c(rnorm(21,0,1),rnorm(21,5,1))
 #hist(data)
 #bimodality_amplitude(data,TRUE)
 #bimodality_coefficient(data,TRUE)
 #bimodality_ratio(data,FALSE)

### 3) Mixture Proportions Examples

 ##Example 3.1
 #dist1<-rnorm(21,5,2)
 #dist2<-dist1+11
 #data<-c(dist1,dist2)
 #hist(data)
 #bimodality_amplitude(data,TRUE)
 #bimodality_ratio(data,FALSE)

 ##Example 3.2
 #dist1<-rnorm(21,-15,1)
 #dist2<-rep(dist1,3)+30
 #data<-c(dist1,dist2)
 #hist(data)
 #bimodality_amplitude(data,TRUE)
 #bimodality_ratio(data,FALSE)

 ##Example 3.4
 #dist1<-rep(7,70)
 #dist2<-rep(-7,70)
 #data<-c(dist1,dist2)
 #hist(data)
 #bimodality_ratio(data,FALSE)


#####		Parametric Examples		#####

### 4) Replicating a two component Gaussian (normal) mixture  
### Example 4.1

 ##Draw data & plot the distribution
 #dist1<-rnorm(14,-5,1)
 #dist2<-rnorm(21,5,1)
 #plot(density(c(dist1,dist2)), main="Bimodal Gaussian mixture distribution")
 
 ##Calculate the means and standard deviations
 #mu1<-mean(dist1)
 #mu2<-mean(dist2)
 #sd1<-sd(dist1)
 #sd2<-sd(dist2)

 ##Apply measures
 #Ashmans_D(mu1,mu2,sd1,sd2)
 #bimodality_separation(mu1,mu2,sd1,sd2)


### 5) Applying to know mixture components  
### Example 5.1

 ##Draw data & plot the distribution
 #data<-c(rnorm(15,0,1),rnorm(21,15,3))
 #plot(density(c(dist1,dist2)), main="Bimodal Gaussian mixture distribution")

 ##Apply measures
 #Ashmans_D(mu1,mu2,sd1,sd2)
 #bimodality_separation(mu1,mu2,sd1,sd2)

sathish-deevi/Roads-Package documentation built on May 29, 2019, 2:59 p.m.