Time-wise autocorrelated and cross-sectionally heteroskedastic disturbances. Using the Nickell, Nunziata and Ochel (2005) data set which is the basis for the empirical example in Sect. 5.3 explaining dynamic unemployment using a panel of 20 OECD countries over the period 1961-95,

(a) Replicate Table 5.8 using iterative generalized least squares allowing for heteroskedastic errors and country- specific first-order serial correlation.

(b) Replicate the rest of Table 5 in Nickell, Nunziata and Ochel (2005, p. 14).

Transcribed Image Text:

Table 5.8 Explaining OECD unemployment, 1961-95 .xtgls ur url ep br bd gd_bd_br c_ud co gd_co_ud tw gd_co_tw lds tfphpc tts d2ms rirl t66-195 tre1-tre20 id2-id20, p(hetero) corr(psar1) rhotype(theil) igls Cross-sectional time-series FGLS regression Coefficients: generalized least squares Panels: Correlation: heteroskedastic panel-specific AR (1) Estimated covariances Estimated autocorrelations Estimated coefficients 228 20 Number of obs 599 20 Number of groups 20 85 Obs per group: min = 12 avg = 29.95 max= 33 Wald chi2 (84) Prob chi2 = 47272.06 = 0.0000