Page **4** of **13**

### New 2nd Edition - The Complete Guide to Option Pricing Formulas + CD-ROM by Espen Gaarder Haug

Posted: **January 12th, 2007, 4:43 pm**

by **Collector**

American put S=100, X=100, T=5, r=0.1, b=0.01, v=0.2CN n=50, m=50 P= 12.0852 CN n=1000, m=1000 p= 12.1822 (I think right at 2 decimals...)500 time step Leisen-Reimer binomial gives 12.1836 (n=50 12.1371)500 CRR tree 12.1841 (N=50 12.1338)500 Trinomial (CRR-equivalent parameters) 12.1838 (n=50 12.1523)

### New 2nd Edition - The Complete Guide to Option Pricing Formulas + CD-ROM by Espen Gaarder Haug

Posted: **January 12th, 2007, 5:28 pm**

by **hgeorgako**

Thank you all for the clarifications.I realize now that the code only keeps track of the last two time steps and not the entire grid. I will adjust the declaration to take into account the C(0 to 1, 0 to M).How about any boundary conditions within the code? The values that i used are same as the ones you outlined but with T = 3. I'm referencing the Wilmott Black Scholes Option Calculator Version 1.0 (

http://paul.wilmott.com/software.cfm)and the CBOE java based calculator here:

http://www.cboe.com/LearnCenter/OptionC ... r.aspxThey both seem to agree.Values for Put = 13.473, Call = 13.473For T = 5 on the Wilmott calculator i get Put = 17.0153. The CBOE calculator only goes up to 3 years. Should i not be trusting these calculators? Any other place where i can cross-reference my results?

### New 2nd Edition - The Complete Guide to Option Pricing Formulas + CD-ROM by Espen Gaarder Haug

Posted: **January 12th, 2007, 5:40 pm**

by **Collector**

I don't know these calculators, are input consistent, what rate, carry, dividend do these calculators require (set to) continuous time? discrete? are you consistently taking into account cost-of-carry/dividend yield... if not sure do not use Black-boxes...it can cost you million of dollars...well also I have to make money I will be on other side

### New 2nd Edition - The Complete Guide to Option Pricing Formulas + CD-ROM by Espen Gaarder Haug

Posted: **January 12th, 2007, 5:41 pm**

by **Collector**

CBOE calculator at T=3 put= 10.582 if you u use rate 10 and dividend 9.... calculator says annual? my guess it is requiering cc rates, that is 10 and 9 (not convereted 10.51709181, 9.417428371) In that case it is fully consitent with my book.... but I have no idea how you got 13.473.....was it you that I sold options to just now at CBOE ? Leisen-Reimer n=500 10.5852CN m=n=1000 10.5824 problem with all such web calculators is that it is often unclear what input they actually require.... once a trader contacted me and told me either my book was wrong or a web based exchange calculator was wrong.... well both where right, I had actually build both of them, but they required different input parameters. Well in this case it was in the instructions...THe CBOE calculator is very unclear in what input is needed...from what they write one should think annual, but continuous compounding seems to be it....?? you can not go into their code so who knows....... I like this: confusion causes opportunities... also in calculators requiering days to expiration, do they then divide by 365? or 365.25....here I think 365? but as long as it is Black-Box it takes time to look into it... my book is no Black-Box if you are putting money to work be careful with any option calculator you use (including my book), no matter.... do not trust your intution....but trust your instincts and know your weapon!!

### New 2nd Edition - The Complete Guide to Option Pricing Formulas + CD-ROM by Espen Gaarder Haug

Posted: **January 12th, 2007, 8:58 pm**

by **Cuchulainn**

QuoteOriginally posted by: CollectorAmerican put S=100, X=100, T=5, r=0.1, b=0.01, v=0.2CN n=50, m=50 P= 12.0852 CN n=1000, m=1000 p= 12.1822 (I think right at 2 decimals...)500 time step Leisen-Reimer binomial gives 12.1836 (n=50 12.1371)500 CRR tree 12.1841 (N=50 12.1338)500 Trinomial (CRR-equivalent parameters) 12.1838 (n=50 12.1523)I tried this on my C++ code. Implicit Euler (first order) and CN (second order)N= UnderlyingM = timeb = r - D so D = 0.09 (i use D in the PDE)//////////////////////////////////////////////////////////1. N = M = 300IE 12.16652CN 12.181482. N = M = 600IE 12.17758CN 12.185113. N = M = 1200IE 12.18304CN 12.18684

### New 2nd Edition - The Complete Guide to Option Pricing Formulas + CD-ROM by Espen Gaarder Haug

Posted: **January 12th, 2007, 9:11 pm**

by **Collector**

n=m=300 12.1693 n=m= 600 12.1785 n=m= 1000 12.1822so looks like you have somewhat faster convergence....but that is as expected I refer to you and a few other FD experts for details on this topic!

### New 2nd Edition - The Complete Guide to Option Pricing Formulas + CD-ROM by Espen Gaarder Haug

Posted: **January 12th, 2007, 9:16 pm**

by **Cuchulainn**

QuoteOriginally posted by: CollectorCBOE calculator at T=3 put= 10.582 if you u use rate 10 and dividend 9.... calculator says annual? my guess it is requiering cc rates, that is 10 and 9 (not convereted 10.51709181, 9.417428371) In that case it is fully consitent with my book.... but I have no idea how you got 13.473.....was it you that I sold options to just now at CBOE ? Leisen-Reimer n=500 10.5852CN m=n=1000 10.5824 With T = 3 I get with N = M = 600IE 10.57983CN 10.5855So I get the same answer as Collector

### New 2nd Edition - The Complete Guide to Option Pricing Formulas + CD-ROM by Espen Gaarder Haug

Posted: **January 12th, 2007, 9:24 pm**

by **Cuchulainn**

QuoteOriginally posted by: Collectorn=m=300 12.1693 n=m= 600 12.1785 n=m= 1000 12.1822so looks like you have somewhat faster convergence....but that is as expected I refer to you and a few other FD experts for details on this topic!We can actually make it even better because this paticular code uses the early exercise test after each time level because this reduces the potential second-order accuracy of CN.But I think that the current values are OK.

### New 2nd Edition - The Complete Guide to Option Pricing Formulas + CD-ROM by Espen Gaarder Haug

Posted: **January 12th, 2007, 9:32 pm**

by **Cuchulainn**

Quoteproblem with all such web calculators is that it is often unclear what input they actually requireCould be a problem of UNITS; some systems allow choose FDM scheme as 0 explicit Euler 1 IE and 0.5 CN while others use 0, 50 and 100!I usually finding the b term confusing when calculating.

### New 2nd Edition - The Complete Guide to Option Pricing Formulas + CD-ROM by Espen Gaarder Haug

Posted: **January 12th, 2007, 10:16 pm**

by **hgeorgako**

Thanks again Gentlemen,Your numbers are indeed correct. The b - term was causing some confusing for me.

### New 2nd Edition - The Complete Guide to Option Pricing Formulas + CD-ROM by Espen Gaarder Haug

Posted: **January 12th, 2007, 10:16 pm**

by **hgeorgako**

Thanks again Gentlemen,Your numbers are indeed correct. The b - term was causing some confusing for me.

### New 2nd Edition - The Complete Guide to Option Pricing Formulas + CD-ROM by Espen Gaarder Haug

Posted: **January 12th, 2007, 10:18 pm**

by **Cuchulainn**

QuoteOriginally posted by: hgeorgakoThanks again Gentlemen,Your numbers are indeed correct. The b - term was causing some confusing for me.You are welcomeYes, I always forget it myself, the Collector book explains all the different variationsI Prefer always b = r - D

### New 2nd Edition - The Complete Guide to Option Pricing Formulas + CD-ROM by Espen Gaarder Haug

Posted: **January 17th, 2007, 3:54 pm**

by **DominicConnor**

It does exist, got my copy yesterday.Looks very good, will take a while...

### New 2nd Edition - The Complete Guide to Option Pricing Formulas + CD-ROM by Espen Gaarder Haug

Posted: **January 22nd, 2007, 4:03 pm**

by **Cuchulainn**

Collector,what kind of CD code you have in book? C#, C++ etc.?

### New 2nd Edition - The Complete Guide to Option Pricing Formulas + CD-ROM by Espen Gaarder Haug

Posted: **January 23rd, 2007, 2:11 am**

by **Collector**

In this book only VBA