@deaneckles Suggested reading: https://t.co/w85pdgxPy0 and also a nice snippet from an email I wrote to Prof. Bell, foreshadowing the current bruhaha about the estimates of criminalization of prostitution in Sweden... https://t.co/sB3IMmAaDL
RT @ArielKarlinsky: @J_ArellanoBover From reading the abstract it seems the original paper heavily relies on age-period-cohort models. I s…
@GoldsteinYoav יש לי חשש שהבעיה כאן היא לא רק טעות בקוד: https://t.co/mTfhCHJrrO
RT @ArielKarlinsky: @J_ArellanoBover From reading the abstract it seems the original paper heavily relies on age-period-cohort models. I s…
RT @ArielKarlinsky: @J_ArellanoBover From reading the abstract it seems the original paper heavily relies on age-period-cohort models. I s…
RT @ArielKarlinsky: @J_ArellanoBover From reading the abstract it seems the original paper heavily relies on age-period-cohort models. I s…
@J_ArellanoBover From reading the abstract it seems the original paper heavily relies on age-period-cohort models. I share the opinion of this paper that such models are fundamentally unidentified: https://t.co/w85pdgxPy0
@ben_golub APC models? really? you can't net them out. https://t.co/w85pdgxPy0
@ikashnitsky Are you of the opinion that APC models are unidentified? I'm pretty convinced after reading Bell & Jones 2017 https://t.co/w85pdgxPy0
RT @AndrewJDBell: @BoulesteixLaure In social sciences, there are lots of age period cohort (APC) examples of this going back decades. More…
@BoulesteixLaure In social sciences, there are lots of age period cohort (APC) examples of this going back decades. More recent ones include the Intrinsic Estimator (https://t.co/D448gMowuM) and the Hierarchical APC model (Table 1 in https://t.co/eVPQGQNUd
RT @MaksimRudnev: A recent summary of debate around age-period-cohort multilevel method from Bell and Jones https://t.co/9z0O82HLuA
A recent summary of debate around age-period-cohort multilevel method from Bell and Jones https://t.co/9z0O82HLuA
@statsmethods So you admit PLS works in only some data scenarios. What’s the criteria for it working? DGP with all 3 of APC (like Y=A+P+C+e)? Data that is ‘real’ & not simulated? Do you think PLS would do better than the HAPC model does here https://t.
RT @AndrewJDBell: My latest on the Hierarchical APC model. TL;DR: it still doesn't work https://t.co/eVPQGQNUd3 https://t.co/xS1igIoxX1
RT @AndrewJDBell: My latest on the Hierarchical APC model. TL;DR: it still doesn't work https://t.co/eVPQGQNUd3 https://t.co/xS1igIoxX1
RT @AndrewJDBell: My latest on the Hierarchical APC model. TL;DR: it still doesn't work https://t.co/eVPQGQNUd3 https://t.co/xS1igIoxX1
ICYMI: @AndrewJDBell takes a look at the Heirarchical APC model. https://t.co/5POVMUjN1H
“mark, what sorts of things do you talk about in your office?” https://t.co/nLOXuq4R5H
RT @AndrewJDBell: My latest on the Hierarchical APC model. TL;DR: it still doesn't work https://t.co/eVPQGQNUd3 https://t.co/xS1igIoxX1
My latest on the Hierarchical APC model. TL;DR: it still doesn't work https://t.co/eVPQGQNUd3 https://t.co/xS1igIoxX1