Regression with part of the base

Asked

Viewed 41 times

-2

I have a base whose one column has three characteristics (strong, weak and moderate. How do I restrict regression so that the model considers only one of the dimensions?

  • 2

    What you mean by "putting a restriction in regression". In regression we do not add restrictions, but variables. What is a dimension? Please edit the question with the answers.

  • Maybe lm(formula, data = subset(base, característica == "forte"))?

1 answer

0

If I understand you right, dataFrame as follows:

> df
   categoria    X   Y
1      Fraco  118  30
2      Fraco  484  58
3      Fraco  664  87
4      Fraco 1004 115
5      Fraco 1231 120
6      Fraco 1372 142
7      Fraco 1582 145
8   Moderado  118  33
9   Moderado  484  69
10  Moderado  664 111
11  Moderado 1004 156
12  Moderado 1231 172
13  Moderado 1372 203
14  Moderado 1582 203
15     Forte  118  30
16     Forte  484  51
17     Forte  664  75
18     Forte 1004 108
19     Forte 1231 115
20     Forte 1372 139
21     Forte 1582 140

So, to apply a regression model to only one category, you just filter the data. For this we have a few options.

1st: filtering the data directly in the model

lm(X~Y, df[df[1] == "Fraco",])

Here I made it df to show only observations in column 1 that are equal to weak.

2nd: Filtering the dataframe with Dplyr

#Carrega o pacote dplyr
library(dplyr)

#filtra a categoria de interesse
df %>% filter(categoria == "Fraco") -> df_fraco

#ajusta o modelo
lm(X~Y, df_fraco)

I hope I’ve helped.

Browser other questions tagged

You are not signed in. Login or sign up in order to post.