, Haoyue Song [aut],
Lixing Zhu [aut]| Title: | The Iterated Score Regression-Based Estimation |
|---|---|
| Description: | We use the ISR to handle with PCA-based missing data with high correlation, and the DISR to handle with distributed PCA-based missing data. The philosophy of the package is described in Guo G. (2024) <doi:10.1080/03610918.2022.2091779>. |
| Authors: | Guangbao Guo [aut, cre]
, Haoyue Song [aut],
Lixing Zhu [aut] |
| Maintainer: | Guangbao Guo <[email protected]> |
| License: | MIT + file LICENSE |
| Version: | 2025.01.14 |
| Built: | 2025-01-27 11:27:19 UTC |
| Source: | https://github.com/cran/ISR |
chronic kidney disease
data("CKD")data("CKD")
The format is: num [1:400, 1:18] 48 7 62 48 51 60 68 24 52 53 ... - attr(*, "dimnames")=List of 2 ..$ : NULL ..$ : chr [1:18] "age" "bp" "sg" "al" ...
There are 1010 missing values in the data set, accounting for 14.03 percent.
Dr.P.Soundarapandian.M.D.,D.M (Senior Consultant Nephrologist), Apollo Hospitals, Managiri, Madurai Main Road, Karaikudi, Tamilnadu, Indi
Polat, H., Danaei-Mehr, H., and Cetin, A. (2017). Diagnosis of chronic kidney disease based on support vector machine by feature selection methods. Journal of Medical Systems, 41(4), 1-11.
data(CKD) ## maybe str(CKD) ; plot(CKD) ...data(CKD) ## maybe str(CKD) ; plot(CKD) ...
Caculate the estimator with the DISR method
DISR(data, data0, real = TRUE, example = FALSE, D)DISR(data, data0, real = TRUE, example = FALSE, D)
data |
is the orignal data set |
data0 |
is the missing data set |
real |
is to judge whether the data set is a real missing data set |
example |
is to judge whether the data set is a simulation example |
D |
is the number of nodes |
XDISR |
is the estimator on the DISR method |
MSEDISR |
is the MSE value of the DISR method |
MAEDISR |
is the MAE value of the DISR method |
REDISR |
is the RE value of the DISR method |
GCVDISR |
is the GCV value of the DISR method |
timeDISR |
is the time cost of the DISR method |
library(MASS) n=100;p=10;per=0.1 X0=data=matrix(mvrnorm(n*p,0,1),n,p) m=round(per*n*p,digits=0) mr=sample(1:(n*p),m,replace=FALSE) X0[mr]=NA;data0=X0 DISR(data=data,data0=data0,real=FALSE,example=FALSE,D=2)library(MASS) n=100;p=10;per=0.1 X0=data=matrix(mvrnorm(n*p,0,1),n,p) m=round(per*n*p,digits=0) mr=sample(1:(n*p),m,replace=FALSE) X0[mr]=NA;data0=X0 DISR(data=data,data0=data0,real=FALSE,example=FALSE,D=2)
Hepatitis C virus
data("HCV")data("HCV")
The format is: num [1:615, 1:13] 1 1 1 1 1 1 1 1 1 1 ... - attr(*, "dimnames")=List of 2 ..$ : chr [1:615] "1" "2" "3" "4" ... ..$ : chr [1:13] "Category" "Age" "Sex" "ALB" ...
There are 31 missing values in the data set, accounting for 0.39 percent.
UCI repository
Lichtinghagen, R., Pietsch, D., Bantel, H., Manns, M., Brand, K. and Bahr, Matthias. (2013). The Enhanced Liver Fibrosis (ELF) Score: Normal Values, Influence Factors and Proposed Cut-Off Values.. Journal of hepatology. 59. 236-242.
data(HCV) ## maybe str(HCV) ; plot(HCV) ...data(HCV) ## maybe str(HCV) ; plot(HCV) ...
Caculate the estimator with the ISR method
ISR(data, data0, real = TRUE, example = FALSE)ISR(data, data0, real = TRUE, example = FALSE)
data |
is the orignal data set |
data0 |
is the missing data set |
real |
is to judge whether the data set is a real missing data set |
example |
is to judge whether the data set is a simulation example. |
XISR |
is the estimator on the ISR method |
MSEISR |
is the MSE value of the ISR method |
MAEISR |
is the MAE value of the ISR method |
REISR |
is the RE value of the ISR method |
GCVISR |
is the GCV value of the ISR method |
timeISR |
is the time cost of the ISR method |
library(MASS) n=100;p=10;per=0.1 X0=data=matrix(mvrnorm(n*p,0,1),n,p) m=round(per*n*p,digits=0) mr=sample(1:(n*p),m,replace=FALSE) X0[mr]=NA;data0=X0 ISR(data=data,data0=data0,real=FALSE,example=FALSE)library(MASS) n=100;p=10;per=0.1 X0=data=matrix(mvrnorm(n*p,0,1),n,p) m=round(per*n*p,digits=0) mr=sample(1:(n*p),m,replace=FALSE) X0[mr]=NA;data0=X0 ISR(data=data,data0=data0,real=FALSE,example=FALSE)
Caculate the estimator on the Mean method
Mean(data, data0, real = TRUE, example = FALSE)Mean(data, data0, real = TRUE, example = FALSE)
data |
is the orignal data set |
data0 |
is the missing data set |
real |
is to judge whether the data set is a real missing data set |
example |
is to judge whether the data set is a simulation example. |
XMean |
is the estimator on the Mean method |
MSEMean |
is the MSE value of the Mean method |
MAEMean |
is the MAE value of the Mean method |
REMean |
is the RE value of the Mean method |
GCVMean |
is the GCV value of the Mean method |
timeMean |
is the time cost of the Mean method |
library(MASS) n=100;p=10;per=0.1 X0=data=matrix(mvrnorm(n*p,0,1),n,p) m=round(per*n*p,digits=0) mr=sample(1:(n*p),m,replace=FALSE) X0[mr]=NA;data0=X0 Mean(data=data,data0=data0,real=FALSE,example=FALSE)library(MASS) n=100;p=10;per=0.1 X0=data=matrix(mvrnorm(n*p,0,1),n,p) m=round(per*n*p,digits=0) mr=sample(1:(n*p),m,replace=FALSE) X0[mr]=NA;data0=X0 Mean(data=data,data0=data0,real=FALSE,example=FALSE)
Caculate the estimator on the MMLPCA method
MMLPCA(data, data0, real = TRUE, example = FALSE)MMLPCA(data, data0, real = TRUE, example = FALSE)
data |
is the orignal data set |
data0 |
is the missing data set |
real |
is to judge whether the data set is a real missing data set |
example |
is to judge whether the data set is a simulation example. |
XMMLPCA |
is the estimator on the MMLPCA method |
MSEMMLPCA |
is the MSE value of the MMLPCA method |
MAEMMLPCA |
is the MAE value of the MMLPCA method |
REMMLPCA |
is the RE value of the MMLPCA method |
GCVMMLPCA |
is the GCV value of the MMLPCA method |
timeMMLPCA |
is the time cost of the MMLPCA method |
library(MASS) n=100;p=10;per=0.1 X0=data=matrix(mvrnorm(n*p,0,1),n,p) m=round(per*n*p,digits=0) mr=sample(1:(n*p),m,replace=FALSE) X0[mr]=NA;data0=X0 MMLPCA(data=data,data0=data0,real=FALSE,example=FALSE)library(MASS) n=100;p=10;per=0.1 X0=data=matrix(mvrnorm(n*p,0,1),n,p) m=round(per*n*p,digits=0) mr=sample(1:(n*p),m,replace=FALSE) X0[mr]=NA;data0=X0 MMLPCA(data=data,data0=data0,real=FALSE,example=FALSE)
Caculate the estimator on the MNIPALS method
MNIPALS(data, data0, real = TRUE, example = FALSE)MNIPALS(data, data0, real = TRUE, example = FALSE)
data |
is the orignal data set |
data0 |
is the missing data set |
real |
is to judge whether the data set is a real missing data set |
example |
is to judge whether the data set is a simulation example. |
XMNIPALS |
is the estimator on the MNIPALS method |
MSEMNIPALS |
is the MSE value of the MNIPALS method |
MAEMNIPALS |
is the MAE value of the MNIPALS method |
REMNIPALS |
is the RE value of the MNIPALS method |
GCVMNIPALS |
is the GCV value of the MNIPALS method |
timeMNIPALS |
is the time cost of the MNIPALS method |
library(MASS) n=100;p=10;per=0.1 X0=data=matrix(mvrnorm(n*p,0,1),n,p) m=round(per*n*p,digits=0) mr=sample(1:(n*p),m,replace=FALSE) X0[mr]=NA;data0=X0 MNIPALS(data=data,data0=data0,real=FALSE,example=FALSE)library(MASS) n=100;p=10;per=0.1 X0=data=matrix(mvrnorm(n*p,0,1),n,p) m=round(per*n*p,digits=0) mr=sample(1:(n*p),m,replace=FALSE) X0[mr]=NA;data0=X0 MNIPALS(data=data,data0=data0,real=FALSE,example=FALSE)
Caculate the estimator on the MRPCA method
MRPCA(data, data0, real = TRUE, example = FALSE)MRPCA(data, data0, real = TRUE, example = FALSE)
data |
is the orignal data set |
data0 |
is the missing data set |
real |
is to judge whether the data set is a real missing data set |
example |
is to judge whether the data set is a simulation example |
XMRPCA |
is the estimator on the MRPCA method |
MSEMRPCA |
is the MSE value of the MRPCA method |
MAEMRPCA |
is the MAE value of the MRPCA method |
REMRPCA |
is the RE value of the MRPCA method |
GCVMRPCA |
is the GCV value of the MRPCA method |
timeMRPCA |
is the time cost of the MRPCA method |
library(MASS) library(MASS) n=100;p=10;per=0.1 X0=data=matrix(mvrnorm(n*p,0,1),n,p) m=round(per*n*p,digits=0) mr=sample(1:(n*p),m,replace=FALSE) X0[mr]=NA;data0=X0 MRPCA(data=data,data0=data0,real=FALSE,example=FALSE)library(MASS) library(MASS) n=100;p=10;per=0.1 X0=data=matrix(mvrnorm(n*p,0,1),n,p) m=round(per*n*p,digits=0) mr=sample(1:(n*p),m,replace=FALSE) X0[mr]=NA;data0=X0 MRPCA(data=data,data0=data0,real=FALSE,example=FALSE)
orange
data("orange")data("orange")
The format is: num [1:12, 1:8] 4.79 4.58 4.71 6.58 NA ... - attr(*, "dimnames")=List of 2 ..$ : chr [1:12] "1" "2" "3" "4" ... ..$ : chr [1:8] "Color.intensity" "Odor.intensity" "Attack.intensity" "Sweet" ...
There are 19 missing values in the data set, accounting for 19.79 percent.
http://factominer.free.fr/missMDA/index.html
Josse J, Husson F (2016). missMDA: A Package for Handling Missing Values in Multivariate Data Analysis. Journal of Statistical Software, 70(1), 1–31.
data(orange) ## maybe str(orange) ; plot(orange) ...data(orange) ## maybe str(orange) ; plot(orange) ...
ozone
data("ozone")data("ozone")
A data frame with 112 observations on the following 11 variables.
maxO3a numeric vector
T9a numeric vector
T12a numeric vector
T15a numeric vector
Ne9a numeric vector
Ne12a numeric vector
Ne15a numeric vector
Vx9a numeric vector
Vx12a numeric vector
Vx15a numeric vector
maxO3va numeric vector
There are 115 missing values in it, accounting for 9.96 percent.
http://factominer.free.fr/missMDA/index.html
Audigier, V., Husson, F., and Josse, J. (2014). A principal components method to impute missing values for mixed data. Advances in Data Analysis and Classification, 10(1), 5-26.
data(ozone) ## maybe str(ozone) ; plot(ozone) ...data(ozone) ## maybe str(ozone) ; plot(ozone) ...
Beijing PM2.5
data("PM2.5")data("PM2.5")
The format is: num [1:43824, 1:12] 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 ... - attr(*, "dimnames")=List of 2 ..$ : chr [1:43824] "1" "2" "3" "4" ... ..$ : chr [1:12] "year" "month" "day" "hour" ...
It records 43824 daily measurements on 12 variables and there are 2067 missing values on 2067 measurements, accounting for 0.00393.
UCI repository
X. Liang, T. Zou, B. Guo, S. Li, H. Zhang, S. Zhang, H. Huang, and S. Chen. Assessing Beijing's PM2.5 pollution: severity, weather impact, APEC and winter heating. Proceedings of the Royal Society A, 471(2182):1–20, 2015.
data(PM2.5) ## maybe str(PM2.5) ; plot(PM2.5) ...data(PM2.5) ## maybe str(PM2.5) ; plot(PM2.5) ...
Travel reviews
data("review")data("review")
The format is: num [1:980, 1:10] 0.93 1.02 1.22 0.45 0.51 0.99 0.9 0.74 1.12 0.7 ... - attr(*, "dimnames")=List of 2 ..$ : chr [1:980] "User_1" "User_2" "User_3" "User_4" ... ..$ : chr [1:10] "Category_1" "Category_2" "Category_3" "Category_4" ...
980 travelers' reviews of 10 different types of travel facilities in East Asia
UCI repository
Renjith, S., Sreekumar, A., and Jathavedan, M. (2018). Evaluation of partitioning clustering algorithms for processing social media data in tourism domain. 2018 IEEE Recent Advances in Intelligent Computational Systems (RAICS), 127-131.
data(review) ## maybe str(review) ; plot(review) ...data(review) ## maybe str(review) ; plot(review) ...
Caculate the estimator on the SR method
SR(data, data0, real = TRUE, example = FALSE)SR(data, data0, real = TRUE, example = FALSE)
data |
is the orignal data set |
data0 |
is the missing data set |
real |
is to judge whether the data set is a real missing data set |
example |
is to judge whether the data set is a simulation example. |
XSR |
is the estimator on the SR method |
MSESR |
is the MSE value of the SR method |
MAESR |
is the MAE value of the SR method |
RESR |
is the RE value of the SR method |
GCVSR |
is the GCV value of the SR method |
library(MASS) n=100;p=10;per=0.1 X0=data=matrix(mvrnorm(n*p,0,1),n,p) m=round(per*n*p,digits=0) mr=sample(1:(n*p),m,replace=FALSE) X0[mr]=NA;data0=X0 SR(data=data,data0=data0,real=FALSE,example=FALSE)library(MASS) n=100;p=10;per=0.1 X0=data=matrix(mvrnorm(n*p,0,1),n,p) m=round(per*n*p,digits=0) mr=sample(1:(n*p),m,replace=FALSE) X0[mr]=NA;data0=X0 SR(data=data,data0=data0,real=FALSE,example=FALSE)