Different results using random forest forecasting in R

Question

When I run an arbitrary forest model from my test data, I get different results for the same model + data model.

Here are the results where you can see the difference in the first column:

> table((predict(rfModelsL[[1]],newdata = a)) ,a$earlyR)

        FALSE TRUE
 FALSE    14    7
 TRUE     13   66

> table((predict(rfModelsL[[1]],newdata = a)) ,a$earlyR)

        FALSE TRUE
 FALSE    15    7
 TRUE     12   66

Although the difference is very small, I am trying to understand what caused this. I assume that it predicthas a “flexible” classification threshold, although I could not find it in the documentation; I'm right?

Thank you in advance

Answer 1

I assume that you have not updated the model here, but it is just a challenge predictthat produces these results. The answer is probably this: from ?predict.randomForest:

, , , ntree randomForest()