1 BINARY CHOICE MODELS: PROBIT ANALYSIS In the case of probit analysis, the sigmoid function is the cumulative standardized normal distribution.
2 BINARY CHOICE MODELS: PROBIT ANALYSIS The maximum likelihood principle is again used to obtain estimates of the parameters.
. probit GRAD ASVABC SM SF MALE Iteration 0: log likelihood = Iteration 1: log likelihood = Iteration 2: log likelihood = Iteration 3: log likelihood = Iteration 4: log likelihood = Probit estimates Number of obs = 540 LR chi2(4) = Prob > chi2 = Log likelihood = Pseudo R2 = GRAD | Coef. Std. Err. z P>|z| [95% Conf. Interval] ASVABC | SM | SF | MALE | _cons | Here is the result of the probit regression using the example of graduating from high school. BINARY CHOICE MODELS: PROBIT ANALYSIS
. probit GRAD ASVABC SM SF MALE Iteration 0: log likelihood = Iteration 1: log likelihood = Iteration 2: log likelihood = Iteration 3: log likelihood = Iteration 4: log likelihood = Probit estimates Number of obs = 540 LR chi2(4) = Prob > chi2 = Log likelihood = Pseudo R2 = GRAD | Coef. Std. Err. z P>|z| [95% Conf. Interval] ASVABC | SM | SF | MALE | _cons | BINARY CHOICE MODELS: PROBIT ANALYSIS As with logit analysis, the coefficients have no direct interpretation. However, we can use them to quantify the marginal effects of the explanatory variables on the probability of graduating from high school.
5 As with logit analysis, the marginal effect of X i on p can be written as the product of the marginal effect of Z on p and the marginal effect of X i on Z. BINARY CHOICE MODELS: PROBIT ANALYSIS
6 The marginal effect of Z on p is given by the standardized normal distribution. The marginal effect of X i on Z is given by i. BINARY CHOICE MODELS: PROBIT ANALYSIS
7 As with logit analysis, the marginal effects vary with Z. A common procedure is to evaluate them for the value of Z given by the sample means of the explanatory variables. BINARY CHOICE MODELS: PROBIT ANALYSIS
8 As with logit analysis, the marginal effects vary with Z. A common procedure is to evaluate them for the value of Z given by the sample means of the explanatory variables. BINARY CHOICE MODELS: PROBIT ANALYSIS. sum GRAD ASVABC SM SF MALE Variable | Obs Mean Std. Dev. Min Max GRAD | ASVABC | SM | SF | MALE |
Probit: Marginal Effects mean b product f(Z) f(Z)b ASVABC SM11.58–0.008– –0.001 SF MALE constant1.00–1.451–1.451 Total BINARY CHOICE MODELS: PROBIT ANALYSIS In this case Z is equal to when the X variables are equal to their sample means.
Probit: Marginal Effects mean b product f(Z) f(Z)b ASVABC SM11.58–0.008– –0.001 SF MALE constant1.00–1.451–1.451 Total BINARY CHOICE MODELS: PROBIT ANALYSIS We then calculate f(Z).
11 BINARY CHOICE MODELS: PROBIT ANALYSIS The estimated marginal effects are f(Z) multiplied by the respective coefficients. We see that a one-point increase in ASVABC increases the probability of graduating from high school by 0.4 percent. Probit: Marginal Effects mean b product f(Z) f(Z)b ASVABC SM11.58–0.008– –0.001 SF MALE constant1.00–1.451–1.451 Total1.881
Probit: Marginal Effects mean b product f(Z) f(Z)b ASVABC SM11.58–0.008– –0.001 SF MALE constant1.00–1.451–1.451 Total Every extra year of schooling of the mother decreases the probability of graduating by 0.1 percent. Father's schooling has no discernible effect. Males have 0.4 percent higher probability than females. BINARY CHOICE MODELS: PROBIT ANALYSIS
Logit Probit Linear f(Z)b f(Z)b b ASVABC SM–0.001–0.001–0.002 SF MALE – The logit and probit results are displayed for comparison. The coefficients in the regressions are very different because different mathematical functions are being fitted. BINARY CHOICE MODELS: PROBIT ANALYSIS
Logit Probit Linear f(Z)b f(Z)b b ASVABC SM–0.001–0.001–0.002 SF MALE – Nevertheless the estimates of the marginal effects are usually similar. BINARY CHOICE MODELS: PROBIT ANALYSIS
Logit Probit Linear f(Z)b f(Z)b b ASVABC SM–0.001–0.001–0.002 SF MALE – However, if the outcomes in the sample are divided between a large majority and a small minority, they can differ. BINARY CHOICE MODELS: PROBIT ANALYSIS
Logit Probit Linear f(Z)b f(Z)b b ASVABC SM–0.001–0.001–0.002 SF MALE – This is because the observations are then concentrated in a tail of the distribution. Although the logit and probit functions share the same sigmoid outline, their tails are somewhat different. BINARY CHOICE MODELS: PROBIT ANALYSIS
Logit Probit Linear f(Z)b f(Z)b b ASVABC SM–0.001–0.001–0.002 SF MALE – BINARY CHOICE MODELS: PROBIT ANALYSIS This is the case here, but even so the estimates are identical to three decimal places. According to a leading authority, Amemiya, there are no compelling grounds for preferring logit to probit or vice versa.
18 Finally, for comparison, the estimates for the corresponding regression using the linear probability model are displayed. BINARY CHOICE MODELS: PROBIT ANALYSIS Logit Probit Linear f(Z)b f(Z)b b ASVABC SM–0.001–0.001–0.002 SF MALE –0.007
Logit Probit Linear f(Z)b f(Z)b b ASVABC SM–0.001–0.001–0.002 SF MALE – If the outcomes are evenly divided, the LPM coefficients are usually similar to those for logit and probit. However, when one outcome dominates, as in this case, they are not very good approximations. BINARY CHOICE MODELS: PROBIT ANALYSIS
Binary Response Models: Interpretation II Probit: g(0)=.4 Logit: g(0)=.25 Linear probability model: g(0)=1 –To make the logit and probit slope estimates comparable, we can multiply the probit estimates by.4/.25=1.6. –The logit slope estimates should be divided by 4 to make them roughly comparable to the LPM (Linear Probability Model) estimates.