This paper discusses the way that theory is used in applied econometri
cs. The traditional strategy of marrying theory and evidence relied on
the fact that older theory implied explicit restrictions on the condi
tional distribution of observable variables and could be evaluated in
terms of the conditional predictions of the model embodying the theore
tical restrictions. However, this is not true of newer theories based
on dynamic stochastic optimisation of models which are not based on qu
adratic objective functions and linear constraints; the so-called 'LQ
form'. Because these models do not usually have closed-form solutions,
they tend to be calibrated rather than estimated and cannot be readil
y evaluated in terms of their conditional predictions. The application
of the stochastic version of the Maximum Principle to such models res
ults in Lagrange multipliers, often shadow prices corresponding to mis
sing markets, which are not observed by the econometrician. Just as ag
ents condition their decisions on unobserved expected prices when forw
ard markets do not exist, they also condition on unobserved shadow pri
ces when particular current or contingent markets do not exist. The ap
proach suggested in this paper is to substitute out the Lagrange multi
pliers in terms of their determinants, just as is often done with expe
ctations. The approach is illustrated in some detail for two examples:
consumer behaviour under liquidity constraints, and oil production.