Is there a spreadsheet available for bootstraping a discount curve from cash, futures and swaps? I have searched the web but can't find anything useful. Most examples use zero coupon or coupon bonds but this is not what is done in practice.

Have a look at QuantLibXL one of the example sheets included in the download does that.

QuantLibXL is a fairly poor learning tool, your not going to get to see how its actually done because everything is hidden behind the formulas.

Yep this was why i asked for spreadsheet in Excel i.e. VBA.I think its something I just have to piece together myself. I am suprised there is something practical like this out there given how fundamental bootstrapping the an interest curve is.I have looked at QuantLib before and found the date classes quite useful but I found the pricing tools not best as a learning tool.

Life just ain't that easy

Hull and Wilmott both have chapters on bootstrapping (I'm pretty sure they both have a worked example).Getting the zero rates from deposits is easy. For futures it's also pretty straightforward (only slight complication is the convexity adjustment) and for swaps you can use the excel solver function. Conventions vary between banks but a typical USD curve would be:O/N, 1w, 1m deposits. futures out to 2 years (say first 6 contracts)swaps for 2 years +of course, there are endless combinations. Depending on how close the near-dated future is you may want to use additional deposit rate(s) (e.g. 3m, 6m) but this mix is close to the Reuters standard curve (they vary the deposits used depending on the near dated future). I would suggest using the quantlibxl functions and data to check your manual efforts, you should be able to get pretty close I would think. To get exact agreement you will need to be consistent on the settlement period, day count, compounding conventions, holidays etc.

what about interpolation methods. always to interpolate the forward curve?

I'm pretty sure you can set the QuantLibXL sheet to do linear interpolation on zeros, which will be good for checking. You can decide on your interpolation method of choice (linear or cubic spline, etc) and your curve - forward curve is one option, or on discount factors is another choice. (I think the Reuters curve uses linear or cubic spline on discount factors.)

May be of use: Lehmans paper

west/hagan paper "Interpolation Methods for Curve Construction" covers interpolation. but linear interpolation of the log of discount factors should be good enough (piecewise constant in instantaneous forward rates).

this topic has been talked about a lot, in the forums..basically, there's a good article to get you started, from the bank of canada.. showing the overall make-up of the curvei can't seem to upload the pdf, here.. go to google and search for 'practical guide to swap curve construction'this shows the treatment of cash, futures(if you choose to use them given your ccy choice) re: liquidity, etc.., then swap ratesfor futures, there are several choices for convexity adjustment.. and then the adjusted futures rate(futures yield - convexity) must be converted to spot rates..then, finally.. the swap rate treatment can be tricky as well..none of my working examples can be uploaded here, but if you email at cvilsack at gmail, i can make available my pdfs..