Estimating the Technology of Cognitive and Noncognitive Skill Formation



functions to be:


gC,j (θC,T+1)  =  μ4,j + α4,C,jθC,T +1

(3.8)


gN,j (θN,T+1) = μ4,j + α4,N,jθN,T +1.

We can also anchor using nonlinear functions. One example would be an outcome pro-
duced by a latent variable
Z4j, for j {J1 + 1,..., J}:

Z4j = gj (θC,T+1,θN,T +1) - ε4,j.

Note that we do not observe Zjj, but we observe the variable Z4,jwhich is defined as:

Z4,j =


1, if gj (θC,T +1, θN,T+1) - ε4,j 0

0, otherwise.

In this notation

Pr ( Z4,j = 1I θC,T+1, θN,T +1) = Pr [ε4,j gj (θC,T+1, θN,T + 1) θC,T +1, θN,T+1]
=
Fε4,j [gj (θC,T +1 ,θN,T +1) θC,T+1N,T +1]
=
gj (θC,T+1, θN,T+1) .

Adult outcomes such as high school graduation, criminal activity, drug use, and teenage
pregnancy may be represented in this fashion.

To establish identification of gj (θC,T +1 , θN,T +1) for j {J1 + 1, . . . , J}, we include the
dummy
Z4,j in the vector θ. Assuming that the dummy Z4,j is measured without error, the
corresponding element of the two repeated measurement vectors
W1 and W2 are identical
and equal to
Z4,j . Theorem 1 implies that the joint density of Z4,j , θC,t and θN,t is identified.
Thus, it is possible to identify Pr [
Z4,j = 1 θC,T+1, θN,T+1].

We can extract two separate “anchors” gC,j (θC,T +1) and gN,j (θN,T +1) from the function
gj (θC,T +1 , θN,T+1), by integrating out the other variable, e.g.,

gC,j (θC,T +1) ≡     gj (θC,T +1, θN,T+1)pθN,T+1 (θN,T+1) N,T +1,            (3.9)

gN,j (θN,T+1) ≡     gj (θC,T +1, θN,T+1)pθC,T+1 (θC,T +1) C,T+1,

where the marginal densities, pθj,T (θN,T +1), j {C, N} are identified by applying the pre-
ceding analysis. Both
gC,j (θC,T +1) and gN,j (θN,T +1) are assumed to be strictly monotonic in

17



More intriguing information

1. Types of Tax Concessions for Promoting Investment in Free Economic and Trade Areas
2. The name is absent
3. Banking Supervision in Integrated Financial Markets: Implications for the EU
4. Credit Market Competition and Capital Regulation
5. Optimal Taxation of Capital Income in Models with Endogenous Fertility
6. The name is absent
7. The name is absent
8. Categorial Grammar and Discourse
9. The quick and the dead: when reaction beats intention
10. Consumer Networks and Firm Reputation: A First Experimental Investigation