Can anyone pls explain base rate neglect with a suitable example. The eg given in CFAI text book is not very clear.
According to wikipedia, “base rate generally refers to the (base) class probabilities unconditioned on featural evidence, frequently also known as prior probabilities.”(https://en.wikipedia.org/wiki/Base_rate) Representative bias is considered placing too much weight on new data. A simple example would be if one has a coin and has a 50/50 odds of heads and tails. If one flips it 2-3 times and it comes up heads, a reprentative bias might ignore the original probablities (base rate) and assume the odds will be more favored to heads(the same as we were getting) or tails( gamblers fallacy) the more interesting example is the coniditional statistics example, such as bayes. we can develop some conditional probabilites (for example, a stock will continue to fall, given it falls). a representative bias of base rate would put too much weight on the new data assuming it will continue to fall ignoring the original probabliies (base rate) which we calculated would happen.
1 Like
Thanks. In other words its the opposite of conservatism bias where base probability is assigned overweight and in base rate neglect we give more weightage to the updated/new information.