Table
3.2
Royal Dutch/Shell Price Differentials and Market MovementsThis table reports regression estimates of the equation:whererRD−SH,is the difference between the log returns of Royal Dutch (Amsterdam) and Shell (London); S&P, FTSE, and DI are returnston the S&P,Financial TimesAllshare index, and Dutch stock indexes, respectively, expressed in their native currencies; andgl/$ andgl/£,represent log changes in the guilder-to-dollar and guilder-to-pound exchange rates. Specification 1 includes leads and lags (shown) toallow for nonsynchronous trading. Specification 2 employs a more restricted set of leads and lags (based on actual time differentials).Specifications 3 and 4 are the same as Specifications 1 and 2, but include a lagged dependent variable on the right-hand side. Durbin’s Al-ternate H (DAH) is reported in place of the Durbin–Watson (DW) statistic for Specifications 3 and 4. Specifications 5, 6, 7, 8 employ 2-,5-, 15-, and 50-day returns. For these specifications, leads and lags of independent variables are dropped. All regressions areOLS, withstandard errors that allow for serial correlation and heteroskedasticity. Where there is only a single coefficient, standard errors are inparentheses.Return DW orLaggedDutchSpecificationHorizonR2DAHDOFDep. Var.S&PFTSEIndexgl/$gl/£1, 1980–19951 day0.2472.3741550.207c−
0.428c0.150c−
0.102c−
0.345c2, 1980–19951 day0.2182.3541640.135c−
0.516c0.365c−
0.123c−
0.612c3, 1980–19951 day0.271−
0.39c4154−
0.174c0.205c−
0.516c0.213c−
0.113c−
0.439c4, 1980–19951 day0.2620.194164−
0.209c0.146c−
0.536c0.359c−
0.121c−
0.612c5, 1980–19952 days0.2042.4219500.064b−
0.451c0.292c−
0.041a−
0.502c(0.032)(0.038)(0.032)(0.030)(0.047)6, 1980–19955 days0.2442.297760.087b−
0.409c0.246c−
0.068a−
0.440c(0.038)(0.042)(0.041)(0.046)(0.070)rSPFTSEDIglgltitiijtj
jktkkltllmtmtmRD SH,&/$/£−+=−+=−+=−+=−+=−=++++++∑∑
∑∑∑
αβδλγυε1 11 11 11 11 1