Ejercicio

library(readxl)
datos_atipicos <- read_excel("C:\\Users\\57321\\Documents\\R\\Excel rstudio\\datos atipicos.xlsx", 
    col_types = c("numeric", "numeric"))
df<-data.frame(datos_atipicos)
summary(df)

##       prot             MO       
##  Min.   :4.761   Min.   :1.820  
##  1st Qu.:5.617   1st Qu.:2.654  
##  Median :5.933   Median :3.072  
##  Mean   :5.886   Mean   :3.033  
##  3rd Qu.:6.223   3rd Qu.:3.439  
##  Max.   :6.966   Max.   :4.072  
##  NA's   :5

library(mice)
df$prot[which(is.na(df$prot))]=mean(df$prot,na.rm = TRUE)

mod_1<-lm(df$prot~df$MO)
summary(mod_1)

## 
## Call:
## lm(formula = df$prot ~ df$MO)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -0.53841 -0.30422  0.05078  0.20115  0.80816 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)   4.1244     0.3337  12.361 3.95e-14 ***
## df$MO         0.5807     0.1082   5.367 5.74e-06 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.3606 on 34 degrees of freedom
## Multiple R-squared:  0.4586, Adjusted R-squared:  0.4427 
## F-statistic:  28.8 on 1 and 34 DF,  p-value: 5.739e-06

#Al hacer el ajuste de los datos con la libreria(mice) a borrar datos extremos e imputar el R cuadrado pasod e 34% a 51,58% acercandolo mas a lo esperado que es 70%.