4 ms·
It's sad to hear that some statisticians don't think they need cross validation. It's not a vapid trend: there are sound statistical reasons to prefer it over o
by nerdponx 10y ago
It's sad to hear that some statisticians don't think they need cross validation. It's not a vapid trend: there are sound statistical reasons to prefer it over other resampling methods like the bootstrap in many applications.