R help - Nov 2010 - lme with time series

Hi:

I'm trying to fit a linear mixed effect model to two time series. My data
base has 3 columns, number of observation, y, and x. Both y and x are market
and specific asset return (both measured on a daily basis for the last 3
years). I want to explain y in terms of x.

Response=y
fixed=intercept
random=x

I've tried lme(y~1,random=~x) and got: Error en
getGroups.data.frame(dataMix, groups) :   Invalid formula for groups

Then I tried lme(rcopec~1,random=~ripsa|obs) (I don´t have groups, each
group is each period of  time and have one observation per group). I got
Error ein n lme.formula(rcopec ~ 1, random = ~ripsa | obs) :
  nlminb problem, convergence error code = 1   message = false convergence
(8)
Warning message:
In lme.formula(rcopec ~ 1, random = ~ripsa | obs) :
  Fewer observations than random effects in all level 1 groups

I can´t make it work. Is there any other function beside lme to do this?

Can some one help me out?

Thanks a lot in advance.

Regards.

María

	[[alternative HTML version deleted]]