### Table 6 Autoregressive distributed lag representation for kpr (standard errors in brackets)

"... In PAGE 19: ... Leaving this question aside for the time being, the starting point for the conditional analysis is an unrestricted autoregressive-distributed lag specification apos;8 for kpr, including the current and two lagged values of the conditioning variables, plus the linear time trend. The OLS estimation results appear in Table6 . While the equation is clearly overparameterized, and individual coefficients are in some cases estimated imprecisely, the long-run solution is well-determined.... In PAGE 19: ... In addition, real credit to the private sector also affects the private capital stock positively, while the relative price of capital has the expected negative effect. The bottom half of Table6 reports a battery of diagnostic tests that reveal no signs of misspecification. According to the test statistics, the residuals show no traces of autocorrelation or auto- regressive conditional heteroskedasticity (ARCH), and Ramsey apos;s RESET test detects no functional form misspecification.... In PAGE 19: ... At most, the Jarque-Bera statistic provides some mild suggestion of non-normality of the residuals. The unrestricted equation in Table6 can be re-arranged and simplified to arrive at an error-correction specification describing both the dynamics of the capital stock and its long-run determinants, along the lines of equation (13) above. This leads to the following model (standard errors in brackets):... In PAGE 20: ...893 SEE = 0.41% Sample: 1962/63-1993/94 The simplifications introduced in equation (15) relative to the GUM model of Table6 are easily accepted by the data (F(4, 16) = 0.... In PAGE 22: ...e., its consistency, as well as the validity of the standard errors and testing procedures reported in Table6 and equation (15) above -- depend crucially on whether the conditioning variables can be taken as weakly exogenous for the parameters of interest. The relevant notion of weak exogeneity depends on the particular choice of apos;parameters of interest apos;.... ..."

### Table 5: Hurricane Andrew Estimation Results for S amp;P Insurer Stock Price Return (Rt) Model

2005

"... In PAGE 11: ... Otherwise, the estimation was undertaken in the same fashion that was used for analyzing the effects of Floyd. Table5 presents the results obtained from estimating the synoptic lifecycle event study for Andrew. Comparing the results of Floyd to Andrew, the former of which ... In PAGE 25: ...Figure 3: Storm Map of Hurricane Andrew Notes: Points on the figure represent the path of the hurricane. Points on this figure correspond to the data presented in Table5 . The path can be differentiated by a variety of storm characteristics.... ..."

### Table 5 Empirical rejection rates under heteroskedastic alternative

"... In PAGE 55: ... - 52 - We turn now, in Table5 , to the power of the IM test against our heteroske- dastic alternative using Monte Carlo critical values. The main feature of these re- sults is that the deviation in the IM statistic generated by our choice of alterna- tives is not enough to be detected reliably in small samples.... In PAGE 58: ... For example, in the probit case with n=50, it yields rejection rates of under 15% for the Chesher-Lancaster form of the statistic in all cases. This is in line with the results in Table5 . Also in line with those results, this procedure generally demonstrates good power with rejection rates that climb toward 100%.... ..."

### Table 1: Percentage Nonsingular and Condition Numbers For Simulated Heteroskedastic Probit Correlation between x1 and x2 0.0 0.10 0.25 0.50 0.75

2006

"... In PAGE 10: ...The results from our Monte Carlo analysis are contained in Table1 . The general result is simple.... In PAGE 11: ... The resulting profile likelihood can be plotted and examined for a lack of 6In particular, we replicated the model that asked respondents whether abortion should be permitted under any circumstances. This is the model in the last column of Table1 in the 1995 AJPS article.... ..."

### Table 1: Unit-Root Tests and Univariate Autoregressive Filters

"... In PAGE 9: ... As the estimated parameters of the conditional mean and variance are asymptotically uncorrelated, this two-stage method is justified. Table1 summarizes the results of the AR filters. DM is the only case for which the pure random walk is adequate for both the spot and futures.... ..."

### Table 6: Variables included in the restricted vector autoregression equations

"... In PAGE 7: ... The final result is a restricted VAR that is generally much simpler than the unrestricted VAR with the same order of lags. While we do not present the estimation results of the final restricted model, we provide Table6 to show the simplification that occurred. Misspecification tests We computed the within-sample residuals of the actual value of the dependent variable less its predicted value and applied various tests on the residuals.... In PAGE 18: ...Table6... ..."

### Table 1: Autoregressions under alternative endpoint speci cationsa

1984

"... In PAGE 9: ... Such VAR speci cations, with all variables included in levels, are typically xed endpoint speci cations as the estimated coe cients are usually consistent with mean-reversion. The top rows of Table1 summarize estimates of univariate twelve-order autoregressions for the one-month bond rate, r, the in ation rate, , and the manufacturing utilization rate, y. The point estimates of the coe cients of the lagged levels of each variable, b, are consistent with mean-reverting behavior.... In PAGE 9: ... Under this speci cation, a long-horizon forecast will converge to the xed endpoint, ?c=b. For the sample spans underlying Table1 , the estimated endpoints are r1 = 6:5, 1 = 4:1, and y1 = 82:8:8 Selected coe cients of a xed endpoints VAR for r, , and y, using a common sample span of 1966m1-1997m99, are shown in the top panel of Table 2.10 The short rate equation plays the role of a policy reaction function where a short-term interest rate, controlled by the policy authority, responds to deviations of in ation from a xed long-run policy target for in ation and to deviations of output from trend.... In PAGE 10: ...2 The moving average endpoints VAR The fundamental di erence between the xed endpoints VAR and the moving average endpoints VAR is that the latter incorporates unit root restrictions on the 1-month interest rate and in ation rate, r and . Returning to the top of Table1 , the format of the autoregressions in r, , and y, is the same as that required for an ADF test of unit root behavior. Indeed, the t-ratios of the coe cients of the lagged dependent variable indicate that a unit root hypothesis cannot be rejected for both the short-term interest rate and in ation rate.... In PAGE 12: ... One check of the reasonableness of this interest rate endpoint is that deviations of the one-month bond rate from this endpoint should be stationary if agents apos; bond rate forecasts in a given period are internally consistent. The fourth equation listed in Table1 indicates that the interest rate deviations from the constructed series of nominal rate endpoints reject the hypothesis of unit root behavior at a 99% signi cance level. It is more di cult to obtain direct measurements of long-horizon endpoints for agents apos; forecasts of in ation.... In PAGE 16: ... However, if additional forecast newsletters also revise their endpoint estimates in the same direction, the aggregate endpoint is eventually dominated by revised endpoint estimates, and subscribers abruptly move away from the most conservative newsletters whose endpoints have not yet been revised. The last two autoregressions shown in Table1 provide evidence against the hypothesis that deviations of monthly in ation from either the aggregated calendar time endpoints series or the aggregated \real time quot; endpoints series are I(1).20 Because the purpose of the VAR is to generate agent expectations that approximate information available to markets when the historical data are recorded, the assumption of symmetric access to policy target information implicit in the xed and moving average endpoints models is dropped, and the \real time quot; in ation endpoint is selected as a more realistic estimate of historical market perceptions of the long-run policy target for in ation.... ..."

Cited by 1

### Table 5 Vector Autoregression Quadravariate: Returns, LIQ, Turnover and Dividend Yield

2003

"... In PAGE 16: ... (13) 5 Empirical Results 5.1 Quadravariate VAR: Returns, LIQ, Turnover, and Dividend Yields In Table5 , we present estimation results for both Samples I and II for a quadravariate VAR(1), which includes excess returns, market liquidity, equity market turnover, and div- idend yields in the set of endogenous variables. First, the excess returns display positive autocorrelation, on average across the countries, for both samples, consistent with Harvey (1995).... In PAGE 18: ...Table5 also displays the Cholesky decomposition of the VAR innovation variance-covariance matrix. Re- call, the variance covariance matrix is allowed to differ across liberalization regimes.... In PAGE 19: ... We also present estimates of cer- tain elements of the Cholesky decomposition that describes the contemporaneous covariance between liquidity and return shocks. The other general effects are broadly unchanged from the quadravariate VAR specification presented in Table5 , and are not reported to conserve space. First, for the bivariate specification, the lagged liquidity measure predicts excess returns across both samples considered, consistent with the pooled regression results presented in Ta- ble 4.... In PAGE 19: ... Table 6 also displays several components of the Cholesky decomposition of the VAR inno- vation variance-covariance matrix . The first estimate (c21), associated with the contempo- raneous covariance between return and liquidity shocks, is nearly identical to the estimates presented in Table5 for the full specification. Because of the nature of the Cholesky decom-... In PAGE 38: ...009 0.013 Quadravariate This table presents maximum likelihood estimates for three alternative VAR specifications: a bivariate VAR including excess returns and LIQ; a trivariate VAR (A) including excess return, LIQ, and turnover; and a trivariate VAR (B) including excess returns, LIQ, and the dividend yield, as well as the quadravariate estimates presented in Table5 . As in Table 5, the Official Liberalization indicator is included in all cases as an additional exogenous variable.... ..."

Cited by 5