I have a project in mind that I am working on, but have little idea where to start. I am a relative newcomer to python (about 1 years exp.) and limited knowledge of quant finance.

What I would like to do is simulate possible yield curves going out 1-3 years. To do this, I planned on scraping the US Treasury website for historical data, and deriving historical daily changes in rates and correlations between terms. Then use Monte Carlo simulation to forecast future curves. - I have built a version of this in excel and crystal ball, but would like to code it in python....

On top of this, I would also like to compute how different bond structures perform during those simulations... bullets, vs short term callable bonds, etc.

I am thinking i would need to use PyMC for the simulation... Any ideas on how to effectively price the various bond structures?

Is this whole thing a worthless idea?


You might be able to forecast interest rates using the yield curve itself. I am writing this on the fly so idk where interest rates are at right now but say if the one year US treasury is at 1% then the expected rate for the 2-yr should be 2% ( since you can gain 1% for one year and at maturity purchase another one year treasury and gain another 1%)... But as you may have notice this is not the case. The one year treasury can be at 1% and the 2-year may be at 2.50% , thus implying that the markets expect an increase in interest rates in the short term since the rate is higher than the "expected" rate of 2%... I hope it helps


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.