Skip to main content

Advertisement

Table 8 Unbiasedness: \(E_{t}[\Delta S_{i,t+k}]=\beta _{0}+\beta _{1}\Delta S_{t+k}+\beta _{2}\lambda (\frac{x'_{it}\beta }{\sigma })+\sum _{j}{\gamma _jD_{year}}+\alpha _i+\epsilon _{it}\)

From: Great expectations? evidence from Colombia’s exchange rate survey

Coefficient/test k = 1 Month k = 1 Year
First differences Fixed effects First differences Fixed effects
\(\beta _{0}\) 0.00 (0.001) −0.00*** (0.001) −0.00 (0.003) 0.07*** (0.005)
\(\beta _{1}\) 0.30*** (0.011) 0.27*** (0.011) 0.26*** (0.022) 0.12*** (0.018)
\(\beta _{2}\) 0.00 (0.003) 0.00 (0.002) 0.00 (0.005) 0.00 (0.005)
\(t: \beta _{1}=1\) 3703*** (0.000) 4750*** (0.000) 1118*** (0.000) 2509*** (0.000)
\(Wald: \beta _{0}=0 \, \beta _{1}=1\) 1857*** (0.000) 2786*** (0.000) 564*** (0.000) 2053*** (0.000)
Observations 3611 4100 2869 3443
  1. Source: authors’ calculations. \(\beta _2\) corresponds to the inverse mills ratio, \(\lambda (\cdot )\), estimated from the attrition probit regression (see Table 3). All estimations were conducted with clustered standard errors, reported in parenthesis. P values are reported only for the t test and Wald test (last two rows). Coefficients for time dummies are not reported. The Hausman test, conducted for all regressions, rejects the null hypothesis in which the unobserved time-invariant component is uncorrelated with the model’s covariates
  2. ***, **, * correspond to significance levels of 1, 5 and 10 %, respectively