## Info

FIGURE 15.4 Probit model: (a) given I,, read P, from the ordinate; (b) given P,, read I, from the abscissa.

610 PARTTHREE: TOPICS IN ECONOMETRICS

Probit Estimation with Grouped Data: gprobit

We will use the same data that we used for glogit, which is given in Table 15.4. Since we already have Pi, the relative frequency (the empirical measure of probability) of owning a house at various income levels as shown in Table 15.5, we can use it to obtain Ii from the normal CDF as shown in Table 15.10, or from Figure 15.5.

Once we have the estimated Ii, estimating and 02 is relatively straightforward, as we show shortly. In passing, note that in the language of probit analysis the unobservable utility index Ii is known as the normal equivalent deviate (n.e.d.) or simply normit. Since the n.e.d. or Ii will be negative whenever Pi < 0 . 5, in practice the number 5 is added to the n.e.d. and the result is called a probit.

TABLE 15.10 ESTIMATING THE INDEX i FROM THE STANDARD NORMAL CDF

TABLE 15.10 ESTIMATING THE INDEX i FROM THE STANDARD NORMAL CDF