https://communities.sas.com/t5/SAS-Academy-for-Data-Science/Calculation-of-Out-Of-Bag-OOB-error-in-a-random-forest-Proc/m-p/652097#M845 <P>Re: Predicting Analytics on Big Data</P> <P>How is the Out-Of-Bag Error calculated for a random forest fitted through RANDOMWOODS in Proc IMSTAT (see page 3-123 of the course text)?</P> <P>Is it an average of the errors on each out-of-bag sample, calculated as Average Square Error?</P> Sun, 31 May 2020 16:30:15 GMT pvareschi 2020-05-31T16:30:15Z https://communities.sas.com/t5/SAS-Academy-for-Data-Science/Calculation-of-Out-Of-Bag-OOB-error-in-a-random-forest-Proc/m-p/652097#M845 <P>Re: Predicting Analytics on Big Data</P> <P>How is the Out-Of-Bag Error calculated for a random forest fitted through RANDOMWOODS in Proc IMSTAT (see page 3-123 of the course text)?</P> <P>Is it an average of the errors on each out-of-bag sample, calculated as Average Square Error?</P> Sun, 31 May 2020 16:30:15 GMT https://communities.sas.com/t5/SAS-Academy-for-Data-Science/Calculation-of-Out-Of-Bag-OOB-error-in-a-random-forest-Proc/m-p/652097#M845 pvareschi 2020-05-31T16:30:15Z https://communities.sas.com/t5/SAS-Academy-for-Data-Science/Calculation-of-Out-Of-Bag-OOB-error-in-a-random-forest-Proc/m-p/652374#M859 Yes you are correct. It is the mean of ASE of all the out-of-bag samples. Mon, 01 Jun 2020 21:21:55 GMT https://communities.sas.com/t5/SAS-Academy-for-Data-Science/Calculation-of-Out-Of-Bag-OOB-error-in-a-random-forest-Proc/m-p/652374#M859 gcjfernandez 2020-06-01T21:21:55Z https://communities.sas.com/t5/SAS-Academy-for-Data-Science/Calculation-of-Out-Of-Bag-OOB-error-in-a-random-forest-Proc/m-p/755575#M1039 <P>Out of bag error is simply error computed on samples not seen during training. Out-of-bag estimate for the generalization error is the error rate of the out-of-bag classifier on the training set (compare it with known yi's). In Breiman's original implementation of the <A href="http://net-informations.com/ds/mla/forest.htm" target="_self">random forest</A> algorithm, each tree is trained on about 2/3 of the total training data. As the forest is built, each tree can thus be tested (similar to leave one out cross validation) on the samples not used in building that tree. This is the out of bag error estimate - an internal error estimate of a random forest as it is being constructed.</P><P>&nbsp;</P> Wed, 21 Jul 2021 07:05:26 GMT https://communities.sas.com/t5/SAS-Academy-for-Data-Science/Calculation-of-Out-Of-Bag-OOB-error-in-a-random-forest-Proc/m-p/755575#M1039 lovelmark 2021-07-21T07:05:26Z