### Table 8 Testing Expectation Hypothesis on Structural Models: Long Horizon

### Table 2: Predicting Changes in Long-Horizon In ation Expectations Estimates of (Standard Errors in Parentheses)

"... In PAGE 9: ... This is not surprising as the Federal Reserve was following a di erent policy regime from October 1979 through September 1982 than afterwards.12 The bottow panel of Table2 includes results from estimation of long-rate change equations over the two sample periods for comparison purposes. Results in the top two panels of Table 2 suggest that when long-horizon in ation expectations are above current in ation, market participants revise down their long-horizon in ation expectations.... In PAGE 9: ...12 The bottow panel of Table 2 includes results from estimation of long-rate change equations over the two sample periods for comparison purposes. Results in the top two panels of Table2 suggest that when long-horizon in ation expectations are above current in ation, market participants revise down their long-horizon in ation expectations. Comparison of results in the top panel with those in the middle panel suggest that the presence of non-trivial taxes magni es the result.... ..."

Cited by 2

### Table 5 Long-Horizon Regressions 1952:4-1998:3

2001

"... In PAGE 28: ...Single Equation Regressions Table5 presents the results of regressions of returns over horizons spanning 1 to 24 quarters on lagged variables. The dependent variable in each regression is the H-period log excess return on the S amp;P Composite Index, rt+1 rf;t+1 + ::: + rt+H rf;t+H.... In PAGE 28: ... For each regression the table reports the estimated coe cient on the included explanatory variable(s), the adjusted R2 statistic, and the Newey-West corrected t-statistic for the hypothesis that the coe cient is zero. The top panel of Table5 reports results from regressions of the log return on the S amp;P Composite Index on to the lagged dividend yield. These results are consistent with those obtained elsewhere (for example, Fama and French 1988; Campbell, Lo and MacKinlay 1997).... In PAGE 28: ... Thus, consistent with existing evidence, the dividend yield is a powerful forecaster of long-horizon returns but has little capacity to forecast short-horizon returns. The next three panels of Table5 gives an indication of the forecasting power of other variables for long-horizon returns. Panel 2 shows that the adding the dividend payout ratio to the equation produces results that are very similar to those using just the dividend yield.... In PAGE 28: ... The R2 statistic suggests that these variables have their greatest predictive power at horizons of 3 years or more, explaining about 40% of the variation in returns at a six year horizon. Table5 reveals that the forecasting power of the trend deviation term is concentrated at... In PAGE 29: ... By including all four variables, the model now has forecasting power for returns at every horizon we consider, although the total fraction of variation in long horizon returns that is predicted remains above that of short horizon returns. These results underscore the nding that d cayt is the best univariate predictor of returns at short to intermediate horizons: at a 4 quarter horizon, the R2 from the regression using just d cayt is almost as large as that in the last panel of Table5 obtained using all four variables. How can we understand the relative strengths and weaknesses of d cayt and dt pt at forecasting returns over di erent horizons? One way to understand these di erences is to note that the discount rates in (12) and (13) di er.... In PAGE 31: ...using the single-equation, direct-estimation approach presented in Table5 . In general, the VAR R2 statistics from the model which includes d cayt are considerably higher than those from the model which excludes this variable.... In PAGE 50: ...Table5 : Long-horizon regressions of excess returns on lagged variables using data from 1952:4-1998:3, OLS estimation. The dependent variable is the sum of H log excess return of the S amp;P composite index, rt+1 rf;t+1 + ::: + rt+H rf;t+H.... ..."

Cited by 18

### Table 6 Vector Autoregression of Excess Returns and Implied Long-Horizon R2

2001

"... In PAGE 52: ...Table6 : Sample period is 1952:4 - 1998:3, OLS estimation. Coe cients from vector autoregressions (VARs) of returns, relative bill rate, dividend yield, dividend payout ratio and the trend deviation term.... ..."

Cited by 18

### Table 6. Long Horizon Predictability Projections: Cumulative Future Return on the Price Dividend Ratio

in Abstract

2007

"... In PAGE 20: ... The link between consumption (its lags) and the price-dividend ratio helps discriminate across the two models. Table6 shows linear projections of cumulative n-year-ahead geometric stock returns on the log price dividend ratio, for n = 1, 2, .... ..."

### Table 6. Long Horizon Predictability Projections: Cumulative Future Return on the Price Dividend Ratio

"... In PAGE 20: ... The link between consumption (its lags) and the price-dividend ratio helps discriminate across the two models. Table6 shows linear projections of cumulative n-year-ahead geometric stock returns on the log price dividend ratio, for n = 1, 2, .... ..."

### Table 6. Long-horizon Regression Estimates Null Hypothesis: No Prior Restrictions on Integration Status (1973q2-1994q4)

1997

"... In PAGE 13: ... Predictability of the Swiss Franc rate is robust to the sample, although the OUT/RW statistics indicate predictability at all horizons and the DM(A) at short horizons. In Table6 , we report the extended sample results with p-values tabulated under the null that or follow an unrestricted VAR. While the p-values are somewhat larger for both in-sample and out-of-sample statistics, they are qualitatively very similar to Table 5.... ..."

Cited by 2