[R] question on integrate function & request for consultant

2004-03-05 Thread s viswanath
Dear R helpers, i am using the ecdf(emp. cumulative distribution function,and am getting the following error on daily returns of stock data (about 2000 data points). I am not sure what this means, I would expect the result of the code to be 1 (the integral from -infinity to +infinity of the e

Re: [R] question on integrate function & request for consultant

2004-03-05 Thread Prof Brian Ripley
You are trying to integrate a discontinous function, and the answer is +Inf. So the message is correct. An ECDF is one to the right of the largest data point. I wonder what you really want to do. On Fri, 5 Mar 2004, s viswanath wrote: > Dear R helpers, > > i am using the ecdf(emp. cumulativ