Académique Documents
Professionnel Documents
Culture Documents
GJMBR - B Classification : FOR Code:150507, 150504, JEL Code: G12 , G13, M31
PricingofIndexOptionsUsingBlacksModel
© 2012. Dr. S. K. Mitra.This is a research/review paper, distributed under the terms of the Creative Commons Attribution-
Noncommercial 3.0 Unported License http://creativecommons.org/licenses/by-nc/3.0/), permitting all non-commercial use,
distribution, and reproduction in any medium, provided the original work is properly cited.
Pricing of Index Options Using Black’s Model
Dr. S. K. Mitra
Abstract - Stock index futures sometimes suffer from ‘a address this problem of negative cost of carry in the
negative cost-of-carry’ bias, as future prices of stock index option pricing model by using forward prices instead of
frequently trade less than their theoretical value that include spot prices. He found that actual forward prices also
carrying costs. Since commencement of Nifty future trading in
capture other irregularity faced by market forces in
India, Nifty future always traded below the theoretical prices.
March 2012
addition to the inventory carrying cost. Black’s model is
This distortion of future prices also spills over to option pricing
and increase difference between actual price of Nifty options found useful for valuing options on physical
and the prices calculated using the famous Black-Scholes commodities where negative cost of carry is common.
formula. Fisher Black tried to address the negative cost of In this study we estimated option prices using
carry effect by using forward prices in the option pricing model both Black’s formula and Black-Scholes formula and
instead of spot prices. Black’s model is found useful for compared these theoretical values with actual prices in 89
valuing options on physical commodities where discounted the market and observed that Black formula gives better
value of future price was found to be a better substitute of spot result in comparison to Black-Scholes formula for Nifty
Version I
prices as an input to value options.
options.
In this study the theoretical prices of Nifty options
using both Black Formula and Black-Scholes Formula were II. Literature Survey
compared with actual prices in the market. It was observed
that for valuing Nifty Options, Black Formula had given better A number of studies are available where
I
t is generally assumed that the relation in prices contracts deviate from their respective fair values. The
between the underlying assets and the futures is arbitrage process are sometimes affected by various
maintained by arbitrageurs. If this relation is frictions such as: regulatory restrictions and barriers,
maintained effectively, then investors find these markets transaction costs, risks from the arbitrage process,
examined the mispricing of futures contracts from their probability distribution of the underlying index using the
fair values. They found that a positive or negative Breeden-Litzenberger formula.
persistence in mispricing existed. They used regression Berkman, Brailsford and Frino (2005) used the
analysis and found that the mispricings are a function of FTSE 100 stock index futures contract and found that
the average absolute daily mispricing and the time to there was a small permanent price impact associated
expiration of the futures. Further analysis of the arbitrage with trades in index futures. Their results revealed that
violations led them to conclude that once an arbitrage the initial price reaction is reversed soon. They did not
band was crossed it was less likely for the mispriced find evidence of asymmetry in the price reaction
value to cross the opposite arbitrage band. They did not following large trades in stock index futures, and
provide information on the size of the average arbitrage suggested that the asymmetry documented in previous
profits. studies was specific to equity markets.
March 2012
Stoll and Whaley (1988) studied the impact of Majority of the studies referred above found
similar strategy on the index futures using a forward- evidence of mispricing in both futures and options
contract pricing relationship. Lee and Nayar (1993) have market. It was also found that mispricing in one
studied the S&P 500 options and futures contracts instrument influence pricing of other instrument. Black’s
traded on the Chicago Board Options Exchange formula paves a way to connect mispricing of futures in
90 (CBOE) and Fung and Chan (1994) analyzed trading in option price estimation
the Chicago Mercantile Exchange (CME), respectively to
detect pricing efficiency and detected presence of III. Pricing Of Futures With Cost Of
Version I
mispricing. Carry
Yadav and Pope (1994), studied the UK
The cost of carry is calculated taking the
markets examining futures against the index and
difference between the actual future prices and the spot
reported that the absolute magnitude of mispricing often
value of the underlying asset. The cost of carry concept
Volume XII Issue III
like natural gas, perishable commodities, etc do not options and overpriced put options. The effect can be
follow cost of carry model as such products can not be termed as the low-call-high-put bias in option prices due
stored. In case of financial assets, cost of carry is the to under pricing of futures.
cost of financing the position that includes cost of funds
blocked. In general, the cost of carry is 'positive' as a IV. Black-Scholes Model And Black’s
result of positive interest and storage costs, but in some Modification
situations it can also be negative.
The Black and Scholes (1973) model of options
The main factors that influence differences
pricing, was a significant development in theoretically
between commodity and equity index futures are:
estimating the option pricing problem. The Black–
• There are no costs of storage involved in holding
Scholes model is attractive since it gives a closed-form
equity (depository costs are negligible).
solution for pricing of European options. With the sole
March 2012
• Equity offers a dividend income, which is a negative
exception of volatility measure, all other variables used
cost when stocks are held and positive when stocks
in the model are observed from the market and therefore
are short sold.
the model had contributed to the expansion of the
Therefore, Cost of carry for a financial product is
options markets as a effective pricing technology were
equal to financing cost – dividend income. When the
made available. Though the original model was
future price at market F is greater than Se-rt then a 91
developed for non-dividend paying securities in
strategy of buying the index and selling the future
European type options, the model can be modified for
contracts will earn risk less profits in excess of the risk-
Version I
the pricing other types of options. The Black-Scholes
free rate, similarly when F is less, then a strategy of
formulas for the prices of European Calls (C) and Puts
selling index and buying futures contracts will achieve
(P) for no dividend paying stocks are given below.
financing below the risk-free rate.
Trading in the stock index derivatives has C = S .N (d1 ) − X .e − rt .N (d 2 )
often change with time. The periods of high volatility are governed by the general shifts in cost of producing a
follow immediately after a large change in the level of commodity and the general shifts of demand of the
the original prices and high volatility usually persists for commodity. Shifts in demands and supply due to fresh
some time. When the underlying assumptions are harvesting can create difference between spot and
violated, the use of Black-Scholes formula to compute future price.
options prices may not always be accurate. In the Black-Scholes formula, the term X e-r t
In spite of the violations mentioned above, represents the present value of exercise price
Black-Scholes model is still very widely used, but discounted at risk free rate r for the time to maturity. The
sometimes adjustments are made to account for some expression is based on the premise that the exercise
of the identified inadequacy. price of the option at a future date includes an interest
rate component over the intrinsic value of exercise price.
a) Black’s Formula
March 2012
cost of carry was addressed by using ‘forward prices’ in selling restrictions of underlying stocks. When future
the option pricing model instead of ‘spot prices’. Black price is higher than the spot value plus cost of carry
observed that actual forward prices not only incorporate (i.e.F > S.ert ) one can sell future and buy underlying
cost of carry but also takes into account other stocks, on the other hand when future price is less than
Volume XII Issue III
irregularities in the market. In his proposed model, he the spot value plus cost of carry, (i.e.F < S.ert ) stocks
substituted spot price (S) by the discounted value of can not be short sold on account of short sell
future price (F.e-rt) in the original Black-Scholes model. restrictions. As a result of this restriction in India, future
Black’s model found application in valuing options on prices time and again trade less than their fair values. It
physical commodities where future price is a better was observed that all the futures of Nifty index were
alternative input for valuing options. trading at prices less than their intrinsic values and in
The Call options prices as per Black’s formula many instances the futures were traded even below their
can be observed solving following equation: spot prices.
Data And Analysis
Global Journal of Management and Business Research
=C Fe − rt .N (d1 ) − X .e − rt .N (d 2 ) V.
Call options of Nifty index traded in National
= e − rt [ F .N (d1 ) − X .N (d 2 ) ] Stock Exchange of India from the period 1st July 2008 to
ln ( F / X ) + (σ 2 / 2 ) t
30th June 2011 were collected from website of the
exchange: www.nseindia.com. Similarly, closing prices
Where d1 =
σ t of the Index for the said period were also gathered.
Since only closing prices of the day were available with
ln( F / X ) −( σ2 / 2) t the exchange, inter-day prices could not be compared.
d2= = d1 − σ t The comparison of closing prices can give error for
σ t thinly traded options due to mismatch of timing. For
example, when an option was traded last at 1.00 pm,
In the formula F is the future price of the asset the spot price of the same asset was different than the
and other input parameters are similar to the inputs closing price taken at the close of the day. Thus the
used in the Black-Scholes model. closing price of the asset can not be used to evaluate a
According to Black, future prices provide trade that took place much earlier. With the availability of
valuable information for the market participants who data from the exchange, comparing call option prices
produce, store and sell commodities. The future prices traded at different times with the corresponding spot
observed during the various transaction months, help values at that time were not possible. This effect of
the producers and traders to decide on the best times to timing mismatch was reduced by short listing only highly
plant, harvest, buy and sell the physical commodity. traded options that were traded till last few minutes and
The future price of a commodity therefore reflects the hence options where the volume of trading was higher
anticipated distribution of Spot prices at the time of than 100 lots in the day were selected for the study.
maturity of the future contract. Black observed that a) Under pricing of Nifty futures
changes in spot price and change in future price are In this study Nifty futures traded during the
usually correlated. Both spot prices and future prices period July 2008 to June 2011 were analyzed. Out of
743 days observation, it was found that in case of 603 price and calculated value was taken as the error of the
days (80.89% of the days) corresponding futures prices formula. The formula that had produced lowest error
were lower their fair value (i.e. spot price plus cost of was considered better. The error for each observation
carry) (table-1). This bias is bound to influence options was added to obtain total error.
pricing in the options market. The future prices quoting N
lower than their fair value were common during the TE = ∑ en
sample period. n =1
In many days the future prices even quoted
below their corresponding spot values. Out of 743 days, Summarized total error data using both Black-
as many as 271 days (36.47% of the cases) futures Scholes model and Black model is presented in table-3.
prices were below their corresponding spot prices. It can be seen that total error in Black model was less
March 2012
than that of B-S model.
b) Other parameters used in option Pricing To find effectiveness of the models, a paired
Other input parameters for estimating call sample t-test was carried out involving error measures
option prices with the formulas are obtained as follows. of the two methods. The Paired Samples t-test
Among input parameters required in the models, the compares the means of two variables. It computes the
standard deviation (σ) of the returns for the duration of difference between the two variables for each case, and 93
the option are not observable from the market and tests to see if the average difference is significantly
therefore an estimate is required. There is no agreement different from zero. The following hypothesis was tested:
Version I
on the suitable method for estimating standard deviation
• Null: There is no significant difference between the
of the returns series. Further, it is a common observation
means of the two variables.
that ‘σ’ of the price series changes with time. Sometime
• Alternate: There is a significant difference between
the returns remain volatile and this volatility persists for
the means of the two variables
The paper tried to address issues related to 7. Gwilm, O.P, & Buckle M.(1999). Volatility
under-pricing of Nifty options on account of negative Forecasting in the Framework of the Option Expiry
cost of carry in futures market. In this study, 29,724 Cycle. The European Journal of Finance, 5,73-94.
option quotes from 1st July 2008 to 30th June 2011, are 8. Hull, John C., (2004), “Introduction to Futures and
analyzed using both B-S formula and Black formula and Options Markets” Second Edition, Prentice-Hall
found that the Black’s formula produce better alternative India,
than use of Black and Scholes formula.. From the 9. Lee, J.H. and N. Nayar (1993), "A Transactions Data
analysis of errors, it is verified that Black model Analysis of Arbitrage Between Index Options and
produces less error than that of Black-Scholes model Index Future", Journal of Futures Markets, 13(8),
and for that reason use of Black model is more fitting 889-902.
than that of B-S model for valuing Nifty options. 10. MacKinlay, C., and K. Ramaswamy, "Index-Futures
March 2012
Political Economy, 81, 637-654. 12. Sternberg, J.S. (1994), "A Re-examination of Put-Call
3. Black, Fischer (1976), “The pricing of commodity Parity on Index Futures", Journal of Futures Markets,
contracts”, Journal of Financial Economics, 3, 167- 14(1), 801-824.
179. 13. Stoll, H.R. and R. E. Whaley, "Futures and Options
Volume XII Issue III
4. Cornell, B., and K. French, "The Pricing of Stock on Stock Indexes: Economic Purpose, Arbitrage,
Index Futures," Journal of Futures Markets, Spring and Market Structure," The Review of Futures
1983a, pp. 1-14. Markets, Vol. 7 No. 2, 1988, pp. 224-249.
14. Yadav, P.K. and P.F Pope (1990), "Stock Index
5. Fleming, J., B. Ostdiek, and R.E. Whaley (1996),
Futures Arbitrage: International Evidence", Journal
"Trading costs and the relative rates of price
of Futures Markets, 10(6), 573-603.
discovery in stock, futures, and option markets",
15. Varma, J. R. (2002), “Mispricing of Volatility in the
Journal of Futures Markets, 16(4), 353-387.
Indian Index Options Market”, Working Paper No.
6. Fung J.K.W. and K.C. Chan (1994), "On the
2002-04-01, Indian Institute of Management,
Global Journal of Management and Business Research
March 2012
Table 2 : Comparison of Future prices vs. Spot Prices
Version I
2 1-Oct-08 31-Dec-08 60 22 38 36.67%
Table 3 : Comparison of Total Error between Black Scholes model and Black’s model
Total Error
Period No. of
Observation
Sl From To
B-S Model Black's Model
1 1-Jul-08 30-Sep-08 1968 (68,764) (48,551)
96
Table 4 : Paired Samples Statistics
Version I
Method
Paired Differences