Frontiers

IDRS-4
Chanchala Ghadge on Frontier Models

These models are similar to those in regression approach but then the difference lies in the fact that we consider the best unlike regression where we consider averages. So obvious difference comes in the errors which are now combination of two ransom variables. One of them is usual regression error and another one will always be positive(because we are considering the best model).

PS: Topic was interesting, but TVR was forced to speak in between all the time (seeing our helpless faces).

Note: IDRS-3 was by Rupali on some fractional Factorial Designs and real life experiment related to it (curd making). I am still unable to develop a liking or interest for DOE. So I don't have to regret much about missing the session. But may be this session was interesting enough to capture me.