Seems excessive to me. Usually implied vol calcs converge in relatively few iterations with a simple NR solver. Given the accuracy and precision of the option prices, I wouldn't ask for such precision in the implieds. Chances are you have some bad prices, which is not unusual even in very active markets. You should probably concentrate on the out-of-the-money options, since they usually give more relevant information. (BTW, I looked on Bloomberg for these options, and it says there are no options in WIG20, either on the cash index or on the futures. Are these OTC's? If so, then, somewhat different cautions may apply.) David Reiner -----Original Message----- From: Wojciech ?lusarski [mailto:wojciech.slusarski@gmail.com] Sent: Thursday, February 17, 2005 1:15 PM To: r-sig-finance@stat.math.ethz.ch Subject: [R-sig-finance] Computing implied volatility using fOptions Hello, I have calculated the implied volatility, for the whole history of option quotes on WIG20 stock index on Warsaw Stock Exchange. The thing that is wondering me is that for some particular days I get volatility nearly 0 (e.g. 3.12236893483001e-11). Is it happening because the option was badly priced those thays (in comparison to Black-Scholes price) or is it a problem of the algorithm. I am usin the GBSVolatility() function with settings: tol <- 10^(-10) maxiter <- 100000 Are those values good for that, or should I use some other values. Best regards, Wojtek _______________________________________________ R-sig-finance@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-finance
Computing implied volatility using fOptions
2 messages · davidr@rhotrading.com, Wojciech Slusarski
Those options are traded on Warsaw Stock Exchange since 22.09.2003. I have calculated all historical implied volatilities and try to compare them with volatility forecasts. This is a very flat market, sometimes there are only few transactions, that's why the prices sometimes are really bad and probably that's the reason why I get such strange implied volatilities. Unfourtunately, because of that I have problem with comparison of those implied volatilities with volatilities forecasts, because each observation like that in time series is seriuosly changing the value of any forecast accuracy measure. Below I enclose a link to the page of Warsaw Stock Exchange describing the standard of the options: http://www.gpw.com.pl/gpw_e.asp?cel=e_papiery&k=261&n=26&i=/derivatives/options/options_WIG20 Best regards, Wojtek On Fri, 18 Feb 2005 13:29:33 -0600, davidr@rhotrading.com
<davidr@rhotrading.com> wrote:
Seems excessive to me. Usually implied vol calcs converge in relatively few iterations with a simple NR solver. Given the accuracy and precision of the option prices, I wouldn't ask for such precision in the implieds. Chances are you have some bad prices, which is not unusual even in very active markets. You should probably concentrate on the out-of-the-money options, since they usually give more relevant information. (BTW, I looked on Bloomberg for these options, and it says there are no options in WIG20, either on the cash index or on the futures. Are these OTC's? If so, then, somewhat different cautions may apply.) David Reiner -----Original Message----- From: Wojciech ?lusarski [mailto:wojciech.slusarski@gmail.com] Sent: Thursday, February 17, 2005 1:15 PM To: r-sig-finance@stat.math.ethz.ch Subject: [R-sig-finance] Computing implied volatility using fOptions Hello, I have calculated the implied volatility, for the whole history of option quotes on WIG20 stock index on Warsaw Stock Exchange. The thing that is wondering me is that for some particular days I get volatility nearly 0 (e.g. 3.12236893483001e-11). Is it happening because the option was badly priced those thays (in comparison to Black-Scholes price) or is it a problem of the algorithm. I am usin the GBSVolatility() function with settings: tol <- 10^(-10) maxiter <- 100000 Are those values good for that, or should I use some other values. Best regards, Wojtek
_______________________________________________ R-sig-finance@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-finance