Skip to main content

Table 9 News decomposition based on VAR estimations

From: What Goliaths and Davids among Swiss firms tell us about expected returns on Swiss asset markets

 Stock market news Bond market news
 (I)(II)(II) (I)(II)(III)
(95% CI)(0.91,1.71)(0.58,1.11)(0.87,1.65)(95% CI)(0.66,7.34)(0.79,8.20)(0.79,8.26)
(95% CI)(0.01,0.02)(0.01,0.01)(0.01,0.02)(95% CI)(0.00,0.00)(0.00,0.00)(0.00,0.00)
var (NRX)0.481.101.23var (NRX)0.540.580.63
(95% CI)(0.34,0.66)(0.81,1.54)(0.89,1.73)(95% CI)(0.25,1.82)(0.26,1.80)(0.28,1.93)
– 2Cov(NCF,NRR)0.11– 0.000.012Cov(NCF,NRR)
(95% CI)(0.07,0.16)(– 0.03,0.03)(– 0.03,0.04)(95% CI)(0.00,0.00)(0.00,0.01)(0.00,0.01)
– 2Cov(NCF,NRX)– 0.73– 1.02– 1.412Cov(NCF,NRX)– 1.53– 1.86– 1.88
(95% CI)(– 1.04, – 0.41)(– 1.47,– 0.64)(– 2.01,– 0.90)(95% CI)(– 2.27, – 0.95)(– 3.03, – 1.09)(– 2.77, – 1.22)
2*Cov(NRR,NRX)– 0.11– 0.03– 0.042*Cov(NRR,NRX)– 0.00– 0.00– 0.00
(95% CI)(– 0.14, – 0.09)(– 0.06,– 0.01)(– 0.07,-0.00)(95% CI)(– 0.00,– 0.00)(– 0.00,0.00)(– 0.00, – 0.00)
  1. Notes: This table gives the variance decomposition of unexpected excess returns on the Swiss stock (left panel) and bond (right panel) markets into variances and covariances of the three components: news about cash-flows (NCF), real interest rates (NRR), and future excess returns (NRX). These statistics are normalized by the variance of the total stock market return news such that they sum to one. We provide the 95% confidence interval of the statistics after 1000 bootstrap simulations in parentheses. The baseline decomposition is based on the estimation of a vector autoregressive model (lag length 1 month) of VAR systems that always include the excess return on the Swiss stock market as its first element, then follow the short-term real interest rate, changes in the nominal short-term rate, the spread between yields on long-term Swiss government bonds, and a short-rate and the log dividend-price ratio of the Swiss stock market. Column (I) presents the estimates from this VAR system for the sample period from January 1999 to December 2017. Column (II) provides the corresponding results from estimating a VAR that additionally includes the US dividend yield and the US term spread as predictive variables. The sample period remains the same. Finally, column (III) in each panel displays the results when we add Swiss GVD to the US variables and the other Swiss variables