GroupAssignment
Wdenhong Ma and Other
2023-04-16
library("ggplot2")
library("ggdendro")
library(tidyverse)## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.1 ✔ readr 2.1.4
## ✔ forcats 1.0.0 ✔ stringr 1.5.0
## ✔ lubridate 1.9.2 ✔ tibble 3.2.1
## ✔ purrr 1.0.1 ✔ tidyr 1.3.0
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errorslibrary(caret)## Loading required package: lattice
##
## Attaching package: 'caret'
##
## The following object is masked from 'package:purrr':
##
## liftpatientData <- read.csv('C:/Users/u6977/Desktop/Patient.csv')
transplantData <- read.csv('C:/Users/u6977/Desktop/Transplant.csv')
creatineData <- read.csv('C:/Users/u6977/Desktop/SerumCreatinine.csv')
transplantdata2 <- transplantData[!duplicated(transplantData$id),]
# Join patientData and transplantData
Data <- full_join(patientData,transplantdata2, by = "id")
# Extract some interesting variables
use <- c('gendercode','creatinineatentry','height','weight','smokingcode','cancerever','chroniclungcode','coronaryarterycode','peripheralvascularcode','cerebrovasularcode','diabetescode','graftno','transplantcentrestate','recipientantibodycmvcode','recipientantibodyebvcode','donorsourcecode','donorage','donorgendercode','ischaemia','ageattransplant','hlamismatchesa','hlamismatchesb','hlamismatchesdr','hlamismatchesdq','maxcytotoxicantibodies','currentcytotoxicantibodies','timeondialysis', 'transplantstatus', 'aliveperiod', 'transplantperiod')
Data <- Data[,use]
# Recode some NA
Data[Data==""] = NA
Data[Data=="-"] = NA
library(tidyverse)
library(dplyr)
library(naniar)
vis_miss(Data)
R Markdown
Create a vector qualitative that contains the names of the qualitative (categorical) variables in the dataset. These are the variables that will be used in further analysis.
Subset the original dataset Data using the variable names in the qualitative vector and store this new dataset in qualitative_data.
Create dummy variables for the categorical variables in the qualitative_data dataset using the dummyVars function from the caret package. This function generates a dummy variable matrix, where each categorical variable is replaced by a set of binary (0/1) variables that indicate the presence or absence of each level (category) of the original variable.
Transform the qualitative_data dataset using the generated dummies matrix. This is done using the predict function, which applies the dummy variable transformation to the dataset. The transformed dataset is stored in data_dummies.
Merge the transformed dataset data_dummies with the original dataset qualitative_data using the cbind function. This function combines the two datasets column-wise (side-by-side) and stores the resulting dataset in data_final
In summary, this R code is preparing a dataset by creating dummy variables for categorical variables and then merging the transformed data with the original data. This is a common preprocessing step when working with categorical data in statistical analysis or machine learning.
sort(colSums(is.na(Data)))## gendercode cancerever
## 0 0
## graftno ageattransplant
## 0 0
## timeondialysis transplantstatus
## 0 0
## aliveperiod transplantperiod
## 0 0
## coronaryarterycode peripheralvascularcode
## 4 4
## cerebrovasularcode diabetescode
## 4 4
## chroniclungcode donorsourcecode
## 5 5
## donorage smokingcode
## 13 15
## transplantcentrestate height
## 24 25
## weight recipientantibodycmvcode
## 27 35
## recipientantibodyebvcode hlamismatchesa
## 38 38
## hlamismatchesb hlamismatchesdr
## 38 41
## currentcytotoxicantibodies donorgendercode
## 53 54
## ischaemia maxcytotoxicantibodies
## 59 75
## creatinineatentry hlamismatchesdq
## 85 740Data <- dplyr::select(Data,-hlamismatchesdq)
vis_miss(Data)
Data <- na.omit(Data)
Data <- droplevels(Data)
vis_miss(Data)
Including Plots
You can also embed plots, for example:
qualitative <- c('gendercode','smokingcode','cancerever','chroniclungcode','coronaryarterycode','peripheralvascularcode','cerebrovasularcode','diabetescode','donorsourcecode','donorgendercode','transplantstatus','aliveperiod')
qualitative_data <- Data[, qualitative]
dummies <- dummyVars(~., data = qualitative_data[, qualitative])
# transform the data using the dummy variables
data_dummies <- predict(dummies, newdata = qualitative_data)
# merge the transformed data with the original data
data_final <- cbind(qualitative_data, data_dummies)
# fit the multiple regression model
model2 <- lm(aliveperiod ~ gendercode+smokingcode+cancerever+chroniclungcode+coronaryarterycode+peripheralvascularcode+cerebrovasularcode+diabetescode+donorsourcecode+donorgendercode+transplantstatus+aliveperiod, data = data_final)## Warning in model.matrix.default(mt, mf, contrasts): the response appeared on
## the right-hand side and was dropped## Warning in model.matrix.default(mt, mf, contrasts): problem with term 12 in
## model.matrix: no columns are assignedsummary(model2)##
## Call:
## lm(formula = aliveperiod ~ gendercode + smokingcode + cancerever +
## chroniclungcode + coronaryarterycode + peripheralvascularcode +
## cerebrovasularcode + diabetescode + donorsourcecode + donorgendercode +
## transplantstatus + aliveperiod, data = data_final)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3261.7 -1351.2 -72.8 1158.9 4190.1
##
## Coefficients:
## Estimate Std. Error
## (Intercept) 2130.029 1709.551
## gendercodeM -32.382 144.662
## smokingcodeFormer 319.735 266.157
## smokingcodeNever 244.353 247.895
## smokingcodeUnknown -1816.892 1221.693
## cancereverYes 369.496 215.777
## chroniclungcodeS 574.055 782.230
## chroniclungcodeY 453.925 377.270
## coronaryarterycodeS -770.839 435.829
## coronaryarterycodeY -384.167 254.856
## peripheralvascularcodeS -949.002 613.201
## peripheralvascularcodeY -175.373 328.954
## cerebrovasularcodeS 411.878 659.249
## cerebrovasularcodeY 430.119 416.542
## diabetescodeO -103.225 258.847
## diabetescodeP -738.121 302.968
## diabetescodeQ -950.654 290.455
## donorsourcecodeBrother 511.567 1730.476
## donorsourcecodeBrother-in-law 2459.651 2451.393
## donorsourcecodeCousin 413.327 1795.902
## donorsourcecodeDaughter 55.030 1827.719
## donorsourcecodeDeceased 7.100 1692.329
## donorsourcecodeDirected kidney exchange -1626.317 1887.366
## donorsourcecodeFather 1041.985 1721.498
## donorsourcecodeFather-in-law -243.510 2081.331
## donorsourcecodeFriend 84.581 1746.899
## donorsourcecodeGrandfather 2177.618 2388.473
## donorsourcecodeHusband 1426.172 1743.658
## donorsourcecodeMonzygotic (Identical Twin Boy) 2833.361 2393.245
## donorsourcecodeMonzygotic (Identical Twin Girl) -554.680 2423.839
## donorsourcecodeMother 623.203 1712.093
## donorsourcecodeNiece -151.752 2070.222
## donorsourcecodeNon-directed, waiting list -716.457 1893.909
## donorsourcecodeOther related (Emotionally - specify) 1206.923 2388.468
## donorsourcecodePartner -879.696 1886.853
## donorsourcecodePathological -57.448 1912.964
## donorsourcecodeSister 7.092 1730.140
## donorsourcecodeSister-in-law 1077.809 1961.270
## donorsourcecodeSon 558.100 1828.556
## donorsourcecodeStepfather 1718.439 2387.144
## donorsourcecodeStepmother 1978.980 2398.896
## donorsourcecodeUncle 879.134 1893.736
## donorsourcecodeWife 169.113 1720.105
## donorgendercodeM 102.561 140.368
## transplantstatus 328.077 161.321
## t value Pr(>|t|)
## (Intercept) 1.246 0.21329
## gendercodeM -0.224 0.82295
## smokingcodeFormer 1.201 0.23013
## smokingcodeNever 0.986 0.32469
## smokingcodeUnknown -1.487 0.13752
## cancereverYes 1.712 0.08737 .
## chroniclungcodeS 0.734 0.46333
## chroniclungcodeY 1.203 0.22940
## coronaryarterycodeS -1.769 0.07748 .
## coronaryarterycodeY -1.507 0.13226
## peripheralvascularcodeS -1.548 0.12227
## peripheralvascularcodeY -0.533 0.59416
## cerebrovasularcodeS 0.625 0.53237
## cerebrovasularcodeY 1.033 0.30223
## diabetescodeO -0.399 0.69020
## diabetescodeP -2.436 0.01514 *
## diabetescodeQ -3.273 0.00113 **
## donorsourcecodeBrother 0.296 0.76763
## donorsourcecodeBrother-in-law 1.003 0.31611
## donorsourcecodeCousin 0.230 0.81806
## donorsourcecodeDaughter 0.030 0.97599
## donorsourcecodeDeceased 0.004 0.99665
## donorsourcecodeDirected kidney exchange -0.862 0.38922
## donorsourcecodeFather 0.605 0.54523
## donorsourcecodeFather-in-law -0.117 0.90690
## donorsourcecodeFriend 0.048 0.96140
## donorsourcecodeGrandfather 0.912 0.36230
## donorsourcecodeHusband 0.818 0.41374
## donorsourcecodeMonzygotic (Identical Twin Boy) 1.184 0.23694
## donorsourcecodeMonzygotic (Identical Twin Girl) -0.229 0.81907
## donorsourcecodeMother 0.364 0.71599
## donorsourcecodeNiece -0.073 0.94159
## donorsourcecodeNon-directed, waiting list -0.378 0.70535
## donorsourcecodeOther related (Emotionally - specify) 0.505 0.61353
## donorsourcecodePartner -0.466 0.64123
## donorsourcecodePathological -0.030 0.97605
## donorsourcecodeSister 0.004 0.99673
## donorsourcecodeSister-in-law 0.550 0.58284
## donorsourcecodeSon 0.305 0.76031
## donorsourcecodeStepfather 0.720 0.47190
## donorsourcecodeStepmother 0.825 0.40974
## donorsourcecodeUncle 0.464 0.64266
## donorsourcecodeWife 0.098 0.92172
## donorgendercodeM 0.731 0.46529
## transplantstatus 2.034 0.04244 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1685 on 572 degrees of freedom
## Multiple R-squared: 0.1262, Adjusted R-squared: 0.05901
## F-statistic: 1.878 on 44 and 572 DF, p-value: 0.0007408quantitative <- c('creatinineatentry','height','weight','recipientantibodycmvcode','donorage','ischaemia','ageattransplant','currentcytotoxicantibodies','timeondialysis','aliveperiod','transplantperiod')
quantitative_data <- Data[, quantitative]
## Linear Regression Model
model <- lm(aliveperiod ~creatinineatentry+height+weight+recipientantibodycmvcode+donorage+ischaemia+ageattransplant+currentcytotoxicantibodies+timeondialysis+aliveperiod+transplantperiod, data = quantitative_data)## Warning in model.matrix.default(mt, mf, contrasts): the response appeared on
## the right-hand side and was dropped## Warning in model.matrix.default(mt, mf, contrasts): problem with term 10 in
## model.matrix: no columns are assignedsummary(model)##
## Call:
## lm(formula = aliveperiod ~ creatinineatentry + height + weight +
## recipientantibodycmvcode + donorage + ischaemia + ageattransplant +
## currentcytotoxicantibodies + timeondialysis + aliveperiod +
## transplantperiod, data = quantitative_data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -867.8 -338.4 -189.5 -11.8 5365.0
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1034.39199 366.95916 2.819 0.00498 **
## creatinineatentry -0.06426 0.08452 -0.760 0.44740
## height -1.85167 2.67090 -0.693 0.48840
## weight 1.68572 2.22374 0.758 0.44871
## recipientantibodycmvcodeNot Done -161.07010 813.16826 -0.198 0.84305
## recipientantibodycmvcodePositive -58.46302 70.68190 -0.827 0.40849
## donorage 3.68906 1.86095 1.982 0.04789 *
## ischaemia -0.97740 5.24024 -0.187 0.85210
## ageattransplant -10.45835 2.52005 -4.150 3.8e-05 ***
## currentcytotoxicantibodies 0.94083 1.95900 0.480 0.63122
## timeondialysis 10.74696 7.53804 1.426 0.15447
## transplantperiod 0.90105 0.02036 44.266 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 809.2 on 605 degrees of freedom
## Multiple R-squared: 0.7869, Adjusted R-squared: 0.783
## F-statistic: 203.1 on 11 and 605 DF, p-value: < 2.2e-16Note that the echo = FALSE parameter was added to the
code chunk to prevent printing of the R code that generated the
plot.