Error: "Error in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...) : NA/NaN/Inf in 'y'"

[johnwhat]

Hi, I've actually had problems with using R I've tried to search for an online solution but I still can't fix it.

My error is:

Error in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...) :

NA/NaN/Inf in 'y'

In addition: Warning message:

In storage.mode(v) <- "double" : NAs introduced by coercion



> attach(dataset)

The following objects are masked from dataset (pos = 3):



Age, CGPA, Employment, games, Gender, handwriting, height, relationship, school,

weight

> mod1= lm(Gender~weight+height+games)

This is roughly how my dataset looks like:

> dataset

# A tibble: 160 x 10

Gender Age Employment school CGPA height weight relationship handwriting games

<chr> <dbl> <chr> <chr> <dbl> <dbl> <dbl> <chr> <dbl> <dbl>

1 Male 19 A student Mckl 3.75 175 60 Single 7 20

2 Female 17 A student Mckl 3.16 165 53 Single 7 10

3 Male 19 A student Taylors ~ 4 175 72 Single 5 6

4 Female 19 A student Methodis~ 3.2 160 53 Taken 3 6

5 Female 19 A student MCKL 3 165 48 Taken 5 6

6 Female 18 A student Methodis~ 3.33 164 70 Single 10 8

7 Female 19 A student Methodis~ 3.58 157 41 Single 5 4

8 Male 19 A student MCKL 3.71 167 64 Taken 3 35

9 Male 20 A student MCKL 4 170 70 Single 5 60

10 Female 18 A student MCKL 3 161 45 Taken 7 6

# ... with 150 more rows

[EconomiCurtis]

@johnwhat

Error in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...) :

NA/NaN/Inf in 'y'

There are a couple of ways to get this error, but I think the issue here is you are regressing over a character vector. Gender is characters "Male" or "Female".

You want to convert that into a numerical variable, 1 or 0 for one gender.

Here's a reprex for that kind of error (and for future reference you'll want to pose questions about coding errors like this with a reprex)

library(dplyr)
df <- tibble(
  y = LETTERS[1:3],
  x1 = 1:3,
  x2 = 3:5
)
lm(y ~ x1 + x2, data = df)
#> Warning in storage.mode(v) <- "double": NAs introduced by coercion
#> Error in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 
#>   NA/NaN/Inf in 'y'

^{Created on 2022-03-26 by the reprex package (v2.0.1)}

Do be careful about how you interpret your results! I don't think weight, height or games will do anything to correctly predict or explain one's gender.