why the predict function does have different rows in R

Question

I'm using linear regression to attempt and forecast certain data. Even though there is no fault, it is strange. In order to forecast data2, which only has 1000 rows, I used a linear regression that I generated from data1, which has 4000 rows.

So, in my opinion, the variable's output should only contain 1000 rows. However, the 4000 variables never stop appearing. I have no idea what's wrong.

model = lm(train$latitude train$crashes)

Model, newdata = test, prediction1 = predict

data.frame = csv1

(Id = c(1:1000), Predicted = prediction1)

From this code, I believe it just needs to predict "test" data which has only 1000 rows. But it says that "arguments imply differing number of rows: 1000, 4337" Here the 4337 is from the train data which has 4337 rows.

Answer 1

Alter this sentence:

model = lm(train$latitude train$crashes) to

lm(crashes latitude, data=train) is the model.
The dataset name should ideally only be supplied once in data= when utilising the formula interface, as opposed to being specified for each variable as train$crashes, etc.

The same crashes latitude formula will be used on the new dataset that you have supplied when you execute predict with newdata=.

You should have been alarmed by the following warning I received when attempting to reproduce your problem:

data.frame with the following formulas: train = crashes = rnorm(1000), latitude = sample(1:1000), test = crashes = rnorm(4377), sample(1:4377)

lm(train$crashes train$latitude) is the model.
prediction1 <- predict(model, newdata = test)
Warning message: 'newdata' had 4337 rows but variables found have 1000 rows

Transform data into actionable insights with our Data Analyst Certification – Enroll today!