because all models are misspecified and all data are subject to measurement errors.
No?
Well if the model misspecification can be pinpointed to a family of functionals and if you deduce a prior on the stochastic behavior of the measurement errors. You have something to work with.
Can’t do this unless you know the correct DGP. You probably sounded smart to a dum guy though (i.e. yourself).
Well if the model misspecification can be pinpointed to a family of functionals and if you deduce a prior on the stochastic behavior of the measurement errors. You have something to work with.Can’t do this unless you know the correct DGP. You probably sounded smart to a dum guy though (i.e. yourself).
Well you know OLS was originally utilized say for agricultural purposes. Say for planting corn, it's a deterministic process, nothing stochastic about the location of a seed with respect to a field.
Well if the model misspecification can be pinpointed to a family of functionals and if you deduce a prior on the stochastic behavior of the measurement errors. You have something to work with.Can’t do this unless you know the correct DGP. You probably sounded smart to a dum guy though (i.e. yourself).
Well you know OLS was originally utilized say for agricultural purposes. Say for planting corn, it's a deterministic process, nothing stochastic about the location of a seed with respect to a field.
well? Well you know? Well?
Well if the model misspecification can be pinpointed to a family of functionals and if you deduce a prior on the stochastic behavior of the measurement errors. You have something to work with.Can’t do this unless you know the correct DGP. You probably sounded smart to a dum guy though (i.e. yourself).
Well you know OLS was originally utilized say for agricultural purposes. Say for planting corn, it's a deterministic process, nothing stochastic about the location of a seed with respect to a field.
Outcome is stochastic. There is RAIN and HUNGRY CATERPILLARS in the error term.
Well if the model misspecification can be pinpointed to a family of functionals and if you deduce a prior on the stochastic behavior of the measurement errors. You have something to work with.
Can’t do this unless you know the correct DGP. You probably sounded smart to a dum guy though (i.e. yourself).
Well you know OLS was originally utilized say for agricultural purposes. Say for planting corn, it's a deterministic process, nothing stochastic about the location of a seed with respect to a field.
well? Well you know? Well?
Haha
Well if the model misspecification can be pinpointed to a family of functionals and if you deduce a prior on the stochastic behavior of the measurement errors. You have something to work with.
Can’t do this unless you know the correct DGP. You probably sounded smart to a dum guy though (i.e. yourself).
Well you know OLS was originally utilized say for agricultural purposes. Say for planting corn, it's a deterministic process, nothing stochastic about the location of a seed with respect to a field.
well? Well you know? Well?
Why do you think in old school regression textbooks presented OLS via nonstochastic regressors. It actually makes sense from historical applications.
Well if the model misspecification can be pinpointed to a family of functionals and if you deduce a prior on the stochastic behavior of the measurement errors. You have something to work with.
Can’t do this unless you know the correct DGP. You probably sounded smart to a dum guy though (i.e. yourself).
Well you know OLS was originally utilized say for agricultural purposes. Say for planting corn, it's a deterministic process, nothing stochastic about the location of a seed with respect to a field.
Outcome is stochastic. There is RAIN and HUNGRY CATERPILLARS in the error term.
Have you heard of green houses? You can deterministically control temp, humidity, irrigation, etc. With DEFINITIVE understanding the error process.