library(tidyverse)
## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.4 ✔ readr 2.1.5
## ✔ forcats 1.0.0 ✔ stringr 1.5.1
## ✔ ggplot2 3.5.1 ✔ tibble 3.2.1
## ✔ lubridate 1.9.3 ✔ tidyr 1.3.1
## ✔ purrr 1.0.2
## ── 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 errors
library(readxl)
library("ggplot2")
library(psych)
##
## Adjuntando el paquete: 'psych'
##
## The following objects are masked from 'package:ggplot2':
##
## %+%, alpha
library(tableone)
library(table1)
##
## Adjuntando el paquete: 'table1'
##
## The following objects are masked from 'package:base':
##
## units, units<-
library(dplyr)
options(scipen = 999, digits = 3, encoding = 'UTF-8')
library(descr)
library(expss)
## Cargando paquete requerido: maditr
##
## Use magrittr pipe '%>%' to chain several operations:
## mtcars %>%
## let(mpg_hp = mpg/hp) %>%
## take(mean(mpg_hp), by = am)
##
##
##
## Adjuntando el paquete: 'maditr'
##
## The following objects are masked from 'package:dplyr':
##
## between, coalesce, first, last
##
## The following object is masked from 'package:purrr':
##
## transpose
##
## The following object is masked from 'package:readr':
##
## cols
##
##
## Adjuntando el paquete: 'expss'
##
## The following objects are masked from 'package:stringr':
##
## fixed, regex
##
## The following objects are masked from 'package:dplyr':
##
## compute, contains, na_if, recode, vars, where
##
## The following objects are masked from 'package:purrr':
##
## keep, modify, modify_if, when
##
## The following objects are masked from 'package:tidyr':
##
## contains, nest
##
## The following object is masked from 'package:ggplot2':
##
## vars
library(DescTools)
##
## Adjuntando el paquete: 'DescTools'
##
## The following object is masked from 'package:maditr':
##
## %like%
##
## The following objects are masked from 'package:psych':
##
## AUC, ICC, SD
library(emmeans)
## Welcome to emmeans.
## Caution: You lose important information if you filter this package's results.
## See '? untidy'
library(summarytools)
##
## Adjuntando el paquete: 'summarytools'
##
## The following objects are masked from 'package:descr':
##
## descr, freq
##
## The following objects are masked from 'package:table1':
##
## label, label<-
##
## The following object is masked from 'package:tibble':
##
## view
library(data.table)
##
## Adjuntando el paquete: 'data.table'
##
## The following object is masked from 'package:DescTools':
##
## %like%
##
## The following objects are masked from 'package:expss':
##
## copy, like
##
## The following objects are masked from 'package:maditr':
##
## copy, dcast, let, melt
##
## The following objects are masked from 'package:lubridate':
##
## hour, isoweek, mday, minute, month, quarter, second, wday, week,
## yday, year
##
## The following objects are masked from 'package:dplyr':
##
## between, first, last
##
## The following object is masked from 'package:purrr':
##
## transpose
library(epiR)
## Cargando paquete requerido: survival
## Package epiR 2.0.77 is loaded
## Type help(epi.about) for summary information
## Type browseVignettes(package = 'epiR') to learn how to use epiR for applied epidemiological analyses
library(sjPlot)
## Install package "strengejacke" from GitHub (`devtools::install_github("strengejacke/strengejacke")`) to load all sj-packages at once!
library(survival)
library(survminer)
## Cargando paquete requerido: ggpubr
##
## Adjuntando el paquete: 'ggpubr'
##
## The following object is masked from 'package:expss':
##
## compare_means
##
##
## Adjuntando el paquete: 'survminer'
##
## The following object is masked from 'package:survival':
##
## myeloma
library(pwr)
library(ggpubr)
library(lmtest)
## Warning: package 'lmtest' was built under R version 4.4.3
## Cargando paquete requerido: zoo
##
## Adjuntando el paquete: 'zoo'
##
## The following objects are masked from 'package:data.table':
##
## yearmon, yearqtr
##
## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numeric
library("openxlsx")
## Warning: package 'openxlsx' was built under R version 4.4.3
library("car")
## Cargando paquete requerido: carData
##
## Adjuntando el paquete: 'car'
##
## The following object is masked from 'package:DescTools':
##
## Recode
##
## The following object is masked from 'package:expss':
##
## recode
##
## The following object is masked from 'package:psych':
##
## logit
##
## The following object is masked from 'package:dplyr':
##
## recode
##
## The following object is masked from 'package:purrr':
##
## some
library("lmtest")
library("MASS")
##
## Adjuntando el paquete: 'MASS'
##
## The following object is masked from 'package:dplyr':
##
## select
dm <- read_excel("C:/Users/Administrador/Downloads/dm.xlsx", sheet = 1)
#transformar en factor Sexo, bajaeduc, imc_cat
dm1=dm
dm <- dm %>% mutate(across(c(imc_cat,sexo,bajaeduc), as.factor))
levels(dm$imc_cat)
## [1] "1" "2" "3"
#Utilizar un modelo de regresión lineal simple para analizar la relación entre la glucemia y el índice de masa corporal (medido en forma continua).
modelo1 <- lm(gluc~imc, data = dm)
confint(modelo1)
## 2.5 % 97.5 %
## (Intercept) 77.612 89.424
## imc 0.496 0.903
tab_model(modelo1)
|
|
gluc
|
|
Predictors
|
Estimates
|
CI
|
p
|
|
(Intercept)
|
83.52
|
77.61 – 89.42
|
<0.001
|
|
imc
|
0.70
|
0.50 – 0.90
|
<0.001
|
|
Observations
|
247
|
|
R2 / R2 adjusted
|
0.157 / 0.154
|
summary(modelo1)
##
## Call:
## lm(formula = gluc ~ imc, data = dm)
##
## Residuals:
## Min 1Q Median 3Q Max
## -30.501 -6.351 0.302 6.100 27.002
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 83.518 2.998 27.85 < 0.0000000000000002 ***
## imc 0.699 0.103 6.76 0.000000000099 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 8.91 on 245 degrees of freedom
## Multiple R-squared: 0.157, Adjusted R-squared: 0.154
## F-statistic: 45.7 on 1 and 245 DF, p-value: 0.0000000000995
#ecuación Glucemia = 83.52 + 0.70x IMC
#Construir un modelo de regresión lineal múltiple para evaluar la relación entre glucemia (gluc) e índice de masa corporal (imc) ajustando por las variables sexo y edad.
dm <- dm %>% mutate(sexo1 = factor(sexo, levels = c("0", "1")))
modelo2 <-lm(gluc ~ imc+edad+sexo1, data = dm)
confint(modelo2)
## 2.5 % 97.5 %
## (Intercept) 72.243 87.623
## imc 0.398 0.802
## edad 0.047 0.245
## sexo11 -5.433 -1.009
tab_model(modelo2)
|
|
gluc
|
|
Predictors
|
Estimates
|
CI
|
p
|
|
(Intercept)
|
79.93
|
72.24 – 87.62
|
<0.001
|
|
imc
|
0.60
|
0.40 – 0.80
|
<0.001
|
|
edad
|
0.15
|
0.05 – 0.24
|
0.004
|
|
sexo1 [1]
|
-3.22
|
-5.43 – -1.01
|
0.004
|
|
Observations
|
247
|
|
R2 / R2 adjusted
|
0.220 / 0.211
|
summary(modelo2)
##
## Call:
## lm(formula = gluc ~ imc + edad + sexo1, data = dm)
##
## Residuals:
## Min 1Q Median 3Q Max
## -27.131 -5.779 0.886 6.233 26.335
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 79.9332 3.9040 20.47 < 0.0000000000000002 ***
## imc 0.6000 0.1024 5.86 0.000000015 ***
## edad 0.1459 0.0502 2.91 0.0040 **
## sexo11 -3.2209 1.1228 -2.87 0.0045 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 8.61 on 243 degrees of freedom
## Multiple R-squared: 0.22, Adjusted R-squared: 0.211
## F-statistic: 22.9 on 3 and 243 DF, p-value: 0.000000000000438
#¿Hay un efecto confundidor de la variable “Bajo nivel educativo” (bajaeduc)
#sobre la asociación entre glucemia e índice de masa corporal? Justificar la
#respuesta.
modelo3 <-lm(gluc ~ imc + bajaeduc, data = dm)
confint(modelo3)
## 2.5 % 97.5 %
## (Intercept) 77.609 89.414
## imc 0.468 0.883
## bajaeduc1 -0.965 3.595
tab_model(modelo3)
|
|
gluc
|
|
Predictors
|
Estimates
|
CI
|
p
|
|
(Intercept)
|
83.51
|
77.61 – 89.41
|
<0.001
|
|
imc
|
0.68
|
0.47 – 0.88
|
<0.001
|
|
bajaeduc [1]
|
1.32
|
-0.96 – 3.59
|
0.257
|
|
Observations
|
247
|
|
R2 / R2 adjusted
|
0.162 / 0.155
|
summary(modelo3)
##
## Call:
## lm(formula = gluc ~ imc + bajaeduc, data = dm)
##
## Residuals:
## Min 1Q Median 3Q Max
## -29.783 -6.053 0.298 6.185 26.523
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 83.512 2.997 27.87 < 0.0000000000000002 ***
## imc 0.676 0.105 6.41 0.00000000076 ***
## bajaeduc1 1.315 1.157 1.14 0.26
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 8.91 on 244 degrees of freedom
## Multiple R-squared: 0.162, Adjusted R-squared: 0.155
## F-statistic: 23.5 on 2 and 244 DF, p-value: 0.000000000453
#baja educ no es confundidor
#Existe la hipótesis de que la asociación entre índice de masa corporal (imc) e
#glucemia podría ser distinta según la edad de las personas. Crear una variable
#dicotómica para edad con punto de corte mayor a 50 años. Testear la hipótesis
#utilizando esta nueva variable. Concluir al respecto.
#MODIF DE EFECTO
dm <- dm %>% mutate(edad_cat=as.factor(case_when(edad>50~"viejo",edad <=50~"joven")))
summary(dm$edad_cat)
## joven viejo
## 87 160
modelo4 <-lm(gluc ~ imc+ edad_cat + imc:edad_cat, data = dm)
confint(modelo4)
## 2.5 % 97.5 %
## (Intercept) 70.574 91.329
## imc 0.338 1.079
## edad_catviejo -7.073 18.095
## imc:edad_catviejo -0.510 0.375
tab_model(modelo4)
|
|
gluc
|
|
Predictors
|
Estimates
|
CI
|
p
|
|
(Intercept)
|
80.95
|
70.57 – 91.33
|
<0.001
|
|
imc
|
0.71
|
0.34 – 1.08
|
<0.001
|
|
edad cat [viejo]
|
5.51
|
-7.07 – 18.10
|
0.389
|
|
imc × edad cat [viejo]
|
-0.07
|
-0.51 – 0.37
|
0.764
|
|
Observations
|
247
|
|
R2 / R2 adjusted
|
0.189 / 0.179
|
summary(modelo4)
##
## Call:
## lm(formula = gluc ~ imc + edad_cat + imc:edad_cat, data = dm)
##
## Residuals:
## Min 1Q Median 3Q Max
## -28.209 -5.839 0.459 6.127 26.096
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 80.9512 5.2685 15.37 < 0.0000000000000002 ***
## imc 0.7086 0.1881 3.77 0.00021 ***
## edad_catviejo 5.5112 6.3885 0.86 0.38917
## imc:edad_catviejo -0.0674 0.2244 -0.30 0.76415
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 8.78 on 243 degrees of freedom
## Multiple R-squared: 0.189, Adjusted R-squared: 0.179
## F-statistic: 18.9 on 3 and 243 DF, p-value: 0.0000000000477
#correr un modelo para mayores de 50 y otro para menores de 50
basemay50 <- subset(dm, edad_cat == "viejo")
base_men50<-subset(dm,edad_cat=="joven")
modelo6viejo<- lm(gluc ~ imc, data = basemay50)
confint(modelo6viejo)
## 2.5 % 97.5 %
## (Intercept) 79.305 93.620
## imc 0.399 0.884
tab_model(modelo6viejo)
|
|
gluc
|
|
Predictors
|
Estimates
|
CI
|
p
|
|
(Intercept)
|
86.46
|
79.31 – 93.62
|
<0.001
|
|
imc
|
0.64
|
0.40 – 0.88
|
<0.001
|
|
Observations
|
160
|
|
R2 / R2 adjusted
|
0.147 / 0.142
|
summary(modelo6viejo)
##
## Call:
## lm(formula = gluc ~ imc, data = basemay50)
##
## Residuals:
## Min 1Q Median 3Q Max
## -26.416 -5.962 0.046 6.527 26.096
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 86.462 3.624 23.86 < 0.0000000000000002 ***
## imc 0.641 0.123 5.22 0.00000056 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 8.81 on 158 degrees of freedom
## Multiple R-squared: 0.147, Adjusted R-squared: 0.142
## F-statistic: 27.2 on 1 and 158 DF, p-value: 0.000000557
modelo6joven<- lm(gluc ~ imc, data = base_men50)
confint(modelo6joven)
## 2.5 % 97.5 %
## (Intercept) 70.533 91.37
## imc 0.337 1.08
tab_model(modelo6joven)
|
|
gluc
|
|
Predictors
|
Estimates
|
CI
|
p
|
|
(Intercept)
|
80.95
|
70.53 – 91.37
|
<0.001
|
|
imc
|
0.71
|
0.34 – 1.08
|
<0.001
|
|
Observations
|
87
|
|
R2 / R2 adjusted
|
0.144 / 0.134
|
summary(modelo6joven)
##
## Call:
## lm(formula = gluc ~ imc, data = base_men50)
##
## Residuals:
## Min 1Q Median 3Q Max
## -28.21 -5.36 1.58 6.06 22.33
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 80.951 5.240 15.45 < 0.0000000000000002 ***
## imc 0.709 0.187 3.79 0.00028 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 8.73 on 85 degrees of freedom
## Multiple R-squared: 0.144, Adjusted R-squared: 0.134
## F-statistic: 14.4 on 1 and 85 DF, p-value: 0.000282
modelo5 <- lm(gluc ~ imc+edad+sexo1+ imc: edad_cat, data = dm)
confint(modelo5)
## 2.5 % 97.5 %
## (Intercept) 70.3915 91.669
## imc 0.3674 0.807
## edad -0.0433 0.294
## sexo11 -5.4203 -0.980
## imc:edad_catviejo -0.1176 0.159
tab_model(modelo5)
|
|
gluc
|
|
Predictors
|
Estimates
|
CI
|
p
|
|
(Intercept)
|
81.03
|
70.39 – 91.67
|
<0.001
|
|
imc
|
0.59
|
0.37 – 0.81
|
<0.001
|
|
edad
|
0.13
|
-0.04 – 0.29
|
0.144
|
|
sexo1 [1]
|
-3.20
|
-5.42 – -0.98
|
0.005
|
|
imc × edad catviejo
|
0.02
|
-0.12 – 0.16
|
0.769
|
|
Observations
|
247
|
|
R2 / R2 adjusted
|
0.221 / 0.208
|
summary(modelo5)
##
## Call:
## lm(formula = gluc ~ imc + edad + sexo1 + imc:edad_cat, data = dm)
##
## Residuals:
## Min 1Q Median 3Q Max
## -26.963 -5.771 0.993 6.205 26.049
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 81.0304 5.4009 15.00 < 0.0000000000000002 ***
## imc 0.5871 0.1115 5.26 0.00000031 ***
## edad 0.1254 0.0857 1.46 0.1444
## sexo11 -3.2002 1.1271 -2.84 0.0049 **
## imc:edad_catviejo 0.0207 0.0702 0.29 0.7685
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 8.63 on 242 degrees of freedom
## Multiple R-squared: 0.221, Adjusted R-squared: 0.208
## F-statistic: 17.1 on 4 and 242 DF, p-value: 0.00000000000221
#modelo que incluye imc, edad y sexo en función de edad
modelo6<- lm(gluc ~ imc+edad+sexo1, data = dm)
confint(modelo6)
## 2.5 % 97.5 %
## (Intercept) 72.243 87.623
## imc 0.398 0.802
## edad 0.047 0.245
## sexo11 -5.433 -1.009
tab_model(modelo6)
|
|
gluc
|
|
Predictors
|
Estimates
|
CI
|
p
|
|
(Intercept)
|
79.93
|
72.24 – 87.62
|
<0.001
|
|
imc
|
0.60
|
0.40 – 0.80
|
<0.001
|
|
edad
|
0.15
|
0.05 – 0.24
|
0.004
|
|
sexo1 [1]
|
-3.22
|
-5.43 – -1.01
|
0.004
|
|
Observations
|
247
|
|
R2 / R2 adjusted
|
0.220 / 0.211
|
summary(modelo6)
##
## Call:
## lm(formula = gluc ~ imc + edad + sexo1, data = dm)
##
## Residuals:
## Min 1Q Median 3Q Max
## -27.131 -5.779 0.886 6.233 26.335
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 79.9332 3.9040 20.47 < 0.0000000000000002 ***
## imc 0.6000 0.1024 5.86 0.000000015 ***
## edad 0.1459 0.0502 2.91 0.0040 **
## sexo11 -3.2209 1.1228 -2.87 0.0045 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 8.61 on 243 degrees of freedom
## Multiple R-squared: 0.22, Adjusted R-squared: 0.211
## F-statistic: 22.9 on 3 and 243 DF, p-value: 0.000000000000438
#evaluación de linealidad
#Generación de predichos
predichos <- fitted.values(modelo6)
#Generación de residuos crudos y estandarizados
res <- residuals(modelo6)
#generación de residuos estandarizados:
standres <-rstandard(modelo6)
#gráfico para evaluar linealidad
plot(y=modelo6$residuals,x=modelo6$fitted.values)
abline(h=0)
#gráfico avplots de residuos de variable x en relación a variable y
avPlots(modelo6)
#evaluación de normalidad
densityPlot(modelo6$residuals)
hist(modelo6$residuals)
boxplot(modelo6$residuals)
qqPlot(standres) #se utilizan residuos estandarizados
## [1] 109 19
describe(modelo6$residuals)
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 247 0 8.56 0.89 0.19 8.85 -27.1 26.3 53.5 -0.22 0.18 0.54
shapiro.test(modelo6$residuals)
##
## Shapiro-Wilk normality test
##
## data: modelo6$residuals
## W = 1, p-value = 0.2
#test de Breusch Pagan de homocedasticidad
bptest(modelo6)
##
## studentized Breusch-Pagan test
##
## data: modelo6
## BP = 1, df = 3, p-value = 0.7
#Independencia: Dubin Watson
dwtest(modelo6)
##
## Durbin-Watson test
##
## data: modelo6
## DW = 2, p-value = 0.6
## alternative hypothesis: true autocorrelation is greater than 0
#multicolinealidad
vif(modelo6)
## imc edad sexo1
## 1.05 1.06 1.05
library(GGally)
## Warning: package 'GGally' was built under R version 4.4.3
## Registered S3 method overwritten by 'GGally':
## method from
## +.gg ggplot2
##
## Adjuntando el paquete: 'GGally'
## The following object is masked from 'package:emmeans':
##
## pigs
vars <- dm[, c("gluc", "sexo", "edad", "imc")]
vars$sexo <- as.factor(vars$sexo)
ggpairs(vars, title = "Matriz de correlaciones y dispersión")
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
#outliers
standres <-rstandard(modelo6)
predichos <- modelo6$fitted.values
plot(y=standres, x=predichos)
abline(h=c(-3, 3))
which(abs(standres) > 3)
## 19 109 223
## 19 109 223
standres[standres >3]
## 223
## 3.08
standres[standres < -3]
## 19 109
## -3.17 -3.17
#valores influyentes
leverage <- as.data.frame(hatvalues(modelo6))
cooksd <- cooks.distance(modelo6)
cooksd[cooksd>(4/247)]#n es el tamaño muestral
## 14 19 46 64 109 129 183 196 202 215 223
## 0.0206 0.0736 0.0185 0.0326 0.0309 0.0235 0.0190 0.0426 0.0169 0.0190 0.0335
## 245
## 0.0210
#Leverage
leverage <- hatvalues(modelo6)
#Distancia de Cook
cooksd <- cooks.distance(modelo6)
#Umbral típico para Cook's Distance
umbral_cook <- 4 / length(cooksd)
#Indices de observaciones influyentes
influyentes <- which(cooksd > umbral_cook)
#Crear tabla resumen
tabla_influyentes <- data.frame(
fila = influyentes,
leverage = leverage[influyentes],
distancia_cook = cooksd[influyentes]
)
#Mostrar tabla
print(tabla_influyentes)
## fila leverage distancia_cook
## 14 14 0.0279 0.0206
## 19 19 0.0285 0.0736
## 46 46 0.0142 0.0185
## 64 64 0.0267 0.0326
## 109 109 0.0122 0.0309
## 129 129 0.0192 0.0235
## 183 183 0.0157 0.0190
## 196 196 0.0256 0.0426
## 202 202 0.0186 0.0169
## 215 215 0.0199 0.0190
## 223 223 0.0139 0.0335
## 245 245 0.0197 0.0210
# Calcular leverage (valores de hat)
leverage <- hatvalues(modelo6)
# Calcular distancia de Cook
cooks <- cooks.distance(modelo6)
#Identificar outliers (residuos estandarizados > 3 o < -3)
outlier_indices <- which(abs(standres) > 3)
#Crear un data frame resumen con info relevante de los outliers
outliers_tabla <- data.frame(
fila = outlier_indices,
valor_real = dm[outlier_indices, modelo6$call$formula[[2]]], # Variable dependiente
predicho = predichos[outlier_indices],
residuo_estandarizado = standres[outlier_indices],
leverage = leverage[outlier_indices],
distancia_cook = cooks[outlier_indices]
)
#Ver tabla
print(outliers_tabla)
## fila gluc predicho residuo_estandarizado leverage distancia_cook
## 19 19 78 105 -3.17 0.0285 0.0736
## 109 109 74 101 -3.17 0.0122 0.0309
## 223 223 135 109 3.08 0.0139 0.0335
#Se corre el modelo sin la observación 19 y sin la observación 196
SIN_19 <- dm [-19, ]
confint(modelo6)
## 2.5 % 97.5 %
## (Intercept) 72.243 87.623
## imc 0.398 0.802
## edad 0.047 0.245
## sexo11 -5.433 -1.009
tab_model(modelo6)
|
|
gluc
|
|
Predictors
|
Estimates
|
CI
|
p
|
|
(Intercept)
|
79.93
|
72.24 – 87.62
|
<0.001
|
|
imc
|
0.60
|
0.40 – 0.80
|
<0.001
|
|
edad
|
0.15
|
0.05 – 0.24
|
0.004
|
|
sexo1 [1]
|
-3.22
|
-5.43 – -1.01
|
0.004
|
|
Observations
|
247
|
|
R2 / R2 adjusted
|
0.220 / 0.211
|
summary(modelo6)
##
## Call:
## lm(formula = gluc ~ imc + edad + sexo1, data = dm)
##
## Residuals:
## Min 1Q Median 3Q Max
## -27.131 -5.779 0.886 6.233 26.335
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 79.9332 3.9040 20.47 < 0.0000000000000002 ***
## imc 0.6000 0.1024 5.86 0.000000015 ***
## edad 0.1459 0.0502 2.91 0.0040 **
## sexo11 -3.2209 1.1228 -2.87 0.0045 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 8.61 on 243 degrees of freedom
## Multiple R-squared: 0.22, Adjusted R-squared: 0.211
## F-statistic: 22.9 on 3 and 243 DF, p-value: 0.000000000000438
modelo9<- lm(gluc ~ imc+edad+sexo1, data = SIN_19)
confint(modelo9)
## 2.5 % 97.5 %
## (Intercept) 71.202 86.357
## imc 0.396 0.792
## edad 0.071 0.267
## sexo11 -5.098 -0.743
tab_model(modelo9)
|
|
gluc
|
|
Predictors
|
Estimates
|
CI
|
p
|
|
(Intercept)
|
78.78
|
71.20 – 86.36
|
<0.001
|
|
imc
|
0.59
|
0.40 – 0.79
|
<0.001
|
|
edad
|
0.17
|
0.07 – 0.27
|
0.001
|
|
sexo1 [1]
|
-2.92
|
-5.10 – -0.74
|
0.009
|
|
Observations
|
246
|
|
R2 / R2 adjusted
|
0.231 / 0.221
|
summary(modelo9)
##
## Call:
## lm(formula = gluc ~ imc + edad + sexo1, data = SIN_19)
##
## Residuals:
## Min 1Q Median 3Q Max
## -27.122 -5.740 0.782 6.112 26.466
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 78.7795 3.8470 20.48 < 0.0000000000000002 ***
## imc 0.5942 0.1005 5.91 0.000000011 ***
## edad 0.1690 0.0498 3.40 0.0008 ***
## sexo11 -2.9202 1.1055 -2.64 0.0088 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 8.45 on 242 degrees of freedom
## Multiple R-squared: 0.231, Adjusted R-squared: 0.221
## F-statistic: 24.2 on 3 and 242 DF, p-value: 0.0000000000000964
SIN_196 <- dm [-196, ]
modelo10<- lm(gluc ~ imc+edad+sexo1, data = SIN_196)
confint(modelo10)
## 2.5 % 97.5 %
## (Intercept) 72.5552 87.764
## imc 0.3730 0.774
## edad 0.0605 0.257
## sexo11 -5.5906 -1.209
tab_model(modelo10)
|
|
gluc
|
|
Predictors
|
Estimates
|
CI
|
p
|
|
(Intercept)
|
80.16
|
72.56 – 87.76
|
<0.001
|
|
imc
|
0.57
|
0.37 – 0.77
|
<0.001
|
|
edad
|
0.16
|
0.06 – 0.26
|
0.002
|
|
sexo1 [1]
|
-3.40
|
-5.59 – -1.21
|
0.002
|
|
Observations
|
246
|
|
R2 / R2 adjusted
|
0.226 / 0.216
|
summary(modelo10)
##
## Call:
## lm(formula = gluc ~ imc + edad + sexo1, data = SIN_196)
##
## Residuals:
## Min 1Q Median 3Q Max
## -27.198 -5.766 0.952 5.985 26.357
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 80.1597 3.8605 20.76 < 0.0000000000000002 ***
## imc 0.5734 0.1018 5.63 0.000000049 ***
## edad 0.1588 0.0499 3.18 0.0016 **
## sexo11 -3.3998 1.1122 -3.06 0.0025 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 8.51 on 242 degrees of freedom
## Multiple R-squared: 0.226, Adjusted R-squared: 0.216
## F-statistic: 23.5 on 3 and 242 DF, p-value: 0.000000000000218
#Crear tablas comparativas de los coeficientes con y sin las observaciones 19 y 196
# Extraer coeficientes
coef_ori <- coef(modelo6)
coef_sin <- coef(modelo9)
# Crear tabla comparativa
comparacion <- data.frame(
Coeficiente = names(coef_ori),
Original = coef_ori,
Sin_19 = coef_sin,
Cambio_pct = 100 * (coef_sin - coef_ori) / coef_ori
)
# Mostrar tabla
print(comparacion)
## Coeficiente Original Sin_19 Cambio_pct
## (Intercept) (Intercept) 79.933 78.780 -1.443
## imc imc 0.600 0.594 -0.962
## edad edad 0.146 0.169 15.842
## sexo11 sexo11 -3.221 -2.920 -9.335
# Extraer coeficientes
coef_ori1 <- coef(modelo6)
coef_sin1 <- coef(modelo10)
# Crear tabla comparativa
comparacion1 <- data.frame(
Coeficiente = names(coef_ori),
Original = coef_ori1,
Sin_196 = coef_sin1,
Cambio_pct = 100 * (coef_sin1 - coef_ori1) / coef_ori1
)
# Mostrar tabla
print(comparacion1)
## Coeficiente Original Sin_196 Cambio_pct
## (Intercept) (Intercept) 79.933 80.160 0.283
## imc imc 0.600 0.573 -4.430
## edad edad 0.146 0.159 8.825
## sexo11 sexo11 -3.221 -3.400 5.556