I am currently trying to do a switching regression or selection framework where I account for selection into the public or private (formal/informal) sectors. For this, I require certain exclusion restrictions (which will affect selection into sectors, but not the wages). I was wondering if anyone would know of any tests I can conduct to check if these exclusion restrictions are strong/valid. For example in an Instrumental Variable framework, I can check whether the instruments are weak or not. But are there any tests in the context of selectivity?
Your help and advice is greatly appreciated. For now, the only way I can think of for testing the exclusion restrictions is through an endogenous dummy variable framework for the pooled sample - and I instrument for the public sector dummy with the exclusion restrictions. Do you think this is a sensible approach if no other tests exist in the selection framework?
Your help and advice is greatly appreciated. For now, the only way I can think of for testing the exclusion restrictions is through an endogenous dummy variable framework for the pooled sample - and I instrument for the public sector dummy with the exclusion restrictions. Do you think this is a sensible approach if no other tests exist in the selection framework?
