Expiration Day Effects: Empirical Evidence from Taiwan
Ju, Shiaw-En, Lo, Keng-Hsin, Wang, Kehluh, Journal of Global Business Issues
This study examines the expiration day effects of TAIEX futures and options. Empirical evidences show that abnormal volatility exists on the day before expiration day. Abnormal volume and abnormal return volatility are found in the first fifteen-minute window on the settlement. Price reversal is identified in the period from the last trading hour of the expiration day through the first fifteen minutes of trading on the settlement day. The implied volatility smile on the expiration day is significantly different from that on comparison days. Moreover, the historical volatility is negatively related to the volatility smile. Overall, the expiration effects do exist in the Taiwan market but tend to shift to the opening period of the settlement day possibly due to the market's unique settlement mechanism.
(ProQuest: ... denotes formulae omitted.)
The first stock index derivatives contracts were launched in the U.S. in the 1980s, with a unique advantage to trade or hedge broad market movements that spread quickly to other financial markets all over the world. The main advantages of the index derivatives are low trading cost and no limitations on speculation. Thus, the derivatives can be an effective hedging instrument and have the function of price discovery. However, huge order imbalances and frantic trading on the expiration day may cause the underlying stock price to temporarily deviate from equilibrium, resulting in abnormal returns, higher volatilities, price reversals and abnormal trading volume. Such expiration day effects have drawn the attention of the regulators and researchers.
There have been extensive studies on the expiration day effects in the U.S. derivatives markets. Stoll and Whaley (1986; 1987) found significant price volatility and an abnormally high volume of stock transaction on expirations days. They also found price reversal on expiration days. In June 1987, the Chicago Mercantile Exchange changed the settlement time of S&P 500 index futures and option contracts from the close of a trading day to the opening in an attempt to mitigate concerns about abnormal stock price movement at "triple witching hours". However, Stoll and Whaley (1991) and Hancock (1993) found that this new approach only shifted the expiration day effects to the opening.
The expiration day effects of index derivatives contracts in other countries also attract researchers' attention. Chamberlain, Cheung, and Kwan (1989) examined Toronto Stock Exchange (TSE) 300 index futures and found significant abnormal volume and price effect. Pope and Yadav (1992) investigated the impact of option expiration on underlying stocks in the United Kingdom. Their experiment results provided evidence in price effect but no significant result was found regarding abnormal return volatility. Swidler, Schwartz and Kristiansen (1994) studied the expiration day effects of options on underlying security traded on the Oslo Stock Exchange. They found evidence of downward price pressure on the underlying stocks on expiration days with a rebound of prices on the following day. Schlag (1996) examined expiration day effects of stock index derivatives in Germany and suggests there was a significant increase in trading volume on expiration days but return volatility remained unchanged around the expiration days. Karolyi (1996) studied Nikkei 225 futures contract expirations and concludes that the expiration of the futures results in abnormal trading volume while significant price effect did not exist. Stoll and Whaley (1997) examined expiration day effects of All Ordinaries Share Price Index (SPI) futures traded on the Sydney Futures Exchange. Their investigations indicated that abnormally high trading volume occured near the close on expiration days but no significant price effects were found around expiration days. Corredor, Lechon and Santamaria (2001) analyzed the expiration day effect of the Ibex-35 derivatives listed on the Spanish Equity Derivatives Exchange. Their findings suggested that the expiration of the Ibex-35 index derivatives was associated with an increase in the trading volume of the underlying asset but had no significant price effect on expiration days. Chow, Yung and Zhang (2003) examined the impact of the expiration of Hang Seng Index derivatives traded on the Hong Kong Futures Exchange (HKFE). Their findings indicated that the expiration days might be associated with a negative price effect and some return volatility on the underlying stock market but there was no evidence of abnormal trading volume or price reversal effect. They concluded that the expiration day effects cannot be confirmed in the Hong Kong market. Chou, Chen and Chen (2006) studied the expiration day effects of Taiwan Futures Exchange (TAIFEX) derivatives between 1998 and 2002 and their findings showed no significant expiration day effects in the emerging derivatives market of Taiwan.
In the literature on the expiration day effects, most findings in countries other than the U.S. are similar to that aim at the U.S. market. However, there are no significant expiration day effects in Hong Kong and Taiwan markets. One possible reason for that lies in the settlement price determination procedure. Stoll and Whaley (1997) discussed the merits of single price and average price settlement procedure. Different settlement price procedures may influence trading activities of hedgers and arbitrageurs. The greatest advantage of a single settlement price for hedgers and arbitrageurs is the convergence between cash and future price at the settlement point. This implies that the arbitrageur has no basis risk since underlying stock positions are unwound at the spot market prices that match the derivatives contract settlement price. Under the average price settlement setting, it is difficult for hedgers and arbitrageurs to unwind their positions by buying or selling proper amount of their stock positions during the settlement price determination period. Thus, the basis risk is introduced in the unwinding. Most derivatives markets, such as the U.S., Japan, German and Australia, use the single price as the settlement price of derivatives. As single price settlement may be criticized for being too easy to manipulate, expiration day effects may be more significant in countries adopting single price settlement than in those using average price. In Hong Kong, the HKFE uses the average of index prices at every five minutes over the entire expiration day as the settlement price. Chow et al. (2003) suggested that the absence of expiration day effects in Hong Kong might be due to the basis risk and thus leaded to a lack of arbitrage activity. Their findings imply the settlement procedure seems to play an important role in expiration day effects.
In Taiwan the TAIFEX uses a unique settlement mechanism different from those used in the U.S., Japan, Hong Kong and other markets. The average price is also used to determine the settlement price but the settlement price determination period is shifted to the first fifteen minutes of the settlement day rather than the whole expiration day used in Hong Kong. The settlement day is the day after the expiration day and the settlement prices for the contracts are computed from the first fifteen-minute volumeweighted average of TAIEX on the settlement day. Chou et al. (2006) suggested that the insignificant expiration day effects in Taiwan are probably due to the separation of the settlement day and the expiration day and the average price settlement procedure.
The first contribution of this study is reexamining the expiration day effects in a fast growing emerging market with a unique settlement mechanism. With evolution of the derivatives market in Taiwan, there has been a significant increase of index derivatives trading volume especially for the TAIEX option (TXO) which was launched on Dec. 24, 2001. The trading volume in 2002 was 1,566,446 contracts and it soared to 96,929,940 contracts in 2006, almost 62 times the contracts traded in 2002. The success of TXO moveed the world ranking of TAIFEXT from No. 57 in 1998 to N0.18 in 2005. At the mean time, TXO has also been ranked No. 5 among the world' s top 10 index options. Chou et al. (2006) set their experiment period from 1998 to 2002, during which time, the index derivatives market was still an infant market. With rapid growth of the index derivatives market, more hedgers, arbitrageurs and speculators participated in the market. Therefore, the expiration day effects may become more and more significant with market evolution. The reexamination results can show new evidence for a fast growing market.
The second attribute of this article is extending expiration day effects analysis to implied volatility smile. From the expiration day effects literature, it is clear that stock markets tend to experience a more turbulent period around expiration days. The implied volatility derived from index option market price is a surrogate of market condition, and therefore the implied volatility smile on expiration days should be different from that on nonexpiration days. Surprisingly, most expiration day effects studies pay little attention to it. That is why we then turn to the implied volatility literature for possible linkage with expiration effects.
Since Black and Scholes (1973) published their famous option pricing model, the intriguing anomaly of "volatility smile" has drawn much attention from researchers and practitioners. Under the setting of Black-Scholes model, the volatility parameter should be constant for variant exercise prices through time. However, implied volatility computed from option market prices appear to be different across exercise prices. This puzzling phenomenon in option markets is called volatility smile.
Most researchers trying to explain the volatility smile focused on relaxing the Black-Scholes constant volatility assumption. Cox and Ross (1976) proposed jump and diffusion processes to find explicit option valuation formulas. Emanuel and MacBetch (1982) examined the power of Cox and Ross constant elasticity of variance (CEV) model to explain the cross-sectional distribution of options prices and found that the CEV model did no better than Black-Scholes model. Rubinstein (1994) developed implied binomial trees model with deterministic local volatility structure that can exactly describe the cross-section of option prices. However, Dumas, Fleming and Whaley (1998) showed that parameters in the simple deterministic volatility structure model were unstable through time.
Other attempts to deal with the smile were based on the stochastic volatility framework. Hull and White (1987) showed that the price of a European option was the BlackScholes price integrated over the probability distribution of the average variance during the life of the option. The problem of their model was that a market price of volatility risk was generally needed. Bates (2000) examined the ability of a stochastic model and found that a jump process can improve the model's ability to generate the volatility smile consistent with market prices, but parameters must set to unreasonable values. From the evidence, neither deterministic nor stochastic model can fully explain the volatility smile.
Another avenue of investigation tried to find the determinants of the implied volatility smile. Pena, Rubio, and Serna (1999) reported that transaction costs, market uncertainty, market momentum and time to expiration were the determinants of the curvature of the volatility smile for the Spanish IBEX-35 index options. Bollen and Whaley (2004) examined the relationship between net buying pressure and volatility smile for S&P 500 index and individual stock options and found that changes in implied volatility were directly related to net buying pressure. They argued that market dealers had to raise options premiums to fully satisfy net buying orders for a specific option segment. In order to absorb large net buying pressure from institutional hedging activities, option premium would usually increase in out-of-money index put options. Chan, Cheng and Lung (2006) investigated net buying pressure in the Hong Kong Hang Seng Index options market during the Asian financial crisis. They found that during the late-crisis, pre-crisis, and post-crisis periods was consistent with Bollen and Whaley's buying pressure hypothesis; but during the whole crisis period the net buying pressure did not have a dominant influence on implied volatility change. They suggested that volatility changes were more like to be affected by market expectations in an extremely turbulent period rather than the net buying pressure.
From the implied volatility smile literature, the reasons for volatility smile are far from conclusive. Analyzing volatility smile on expiration may shed lights on the existing literature. Furthermore, the multiple regression is employed to examine whether the determinants affecting volatility smile vary with trading on expiration or not.
The remainder of this paper is organized as follows. The next section presents the data and methodology used in the study. The empirical results are reported and discussed after that, and this is followed by our conclusion.
Data and Methodology
The Taiwan Stock Exchange Capitalization Weighted Stock Index (TAIEX) is a valueweighted index comprising all stocks listed on the Taiwan Stock Exchange. The futures and options on the TAIEX were launched on the TAIFEX on July 21, 1998 and Dec. 24, 2001 respectively. The TAIEX futures and option contracts have a monthly expiration cycle, with the last trading day on the third Wednesday of the delivery month, and the trading calendar comprises the three nearest consecutive months and the other two months of March- June-September-December quarterly cycle. The settlement price for index derivatives contracts is determined from the first fifteen minutes volume-weighted average of TAIEX on the day after expiration day (Thursday). TXO is European, thus it cannot be exercised until expiration.
The data used in this study are drawn from two sources. Intraday TAIEX and trading volume data are collected from Taiwan Stock Exchange statistical reports. Futures and options on TAIEX and related parameters needed in calculating implied volatility are provided by Taiwan Economic Journal (TEJ) database. The experiment period is from Jan. 2002 to Dec. 2006. We employ the comparison-period-approach (CPA) proposed by Masulis (1980) to test whether expiration day effects exist in Taiwan market. The CPA is used to compare the averaged returns, stock price volatility, trading volume, and implied volatility smile on expiration days with those on comparison days. It is possible arbitragers and hedgers offset their positions earlier and the unique settlement mechanism in Taiwan market which separates expiration day and settlement day, therefore we extend our analysis to one day before expiration day (i.e. Tuesday) and settlement day (i.e. Thursday). The control group of comparison days consist of three Wednesdays (Tuesdays, Thursdays) preceding each expiration Wednesday (Tuesday, Thursday).
Abnormal Return Effect
To test the abnormal return effect, ten windows are selected including Tuesday, Wednesday, and Thursday windows; the first 15, 30, 60-minute and the last 15, 30, 60-minute windows on Wednesday, and the first 15-minute hour window on Thursday. Then we define the rate of return on the TAIEX at At intervals during month i as
where P^sub i,t^ is the index level at the beginning of interval t . The average rate of return for the expiration and the comparison days are defined as
... (2) and
where N^sub e^ and N^sub c^ are the number of expiration day and the comparison day, respectively. A t-test is used to examine whether the average returns of the stock market for the expiration days are significantly different from that on the comparison days. The null and alternative hypothesis are H^sub 0^ : AR^sub e^ = AR^sub c^ , H^sub 1^ : AR^sub e^ ≠ AR^sub c^ , respectively.
Price Reversal Effect
We adopt the measure used by Stoll and Whaley (1991) to examine the price reversal effect. Arbitragers may liquidate their positions when the derivative contracts approach expiration day. Speculators may attempt to lower or raise the spot price to increase the profit. All these trading activities will cause spot market price deviate from equilibrium. Nevertheless, this impact may be temporary, when the expiration day passes then the stock prices could reverse in the opposite direction. We first calculate the returns of the 60-minute period before the closing on the expiration day, R^sub t^ , and the return after the closing, R^sub l+1^ , as
... (4) and
respectively, where P^sub close,t^ is the closing index on the expiration day, P^sub close-60,t^ is the index 60 minutes before the market close on the expiration day, and P^sub open,t+1^ is the opening index on the day after expiration day. Then, the first measure of the price reversal is
Because TAIFEX employs the average trading price of the first fifteen minutes of the settlement day (the day after the expiration day) as the settlement price, the expiration day effect might shift to the settlement day. We then define a new measure of return up to settlement instant as
where P^sub open+15,t+1^ is the index for the first fifteen-minute period after the opening on the settlement day. The price reversal measure is then redefined as
As mentioned above, the speculators may attempt to influence the settlement price during the first fifteen trading minutes on the settlement day. Once the settlement price is determined, the price pressure suddenly disappears. Thus we define another set of returns as
... (9) and
where P^sub open,l+1^ and P^sub close,t+1^ are the opening and closing prices, respectively, on the settlement day. We then define the third measure of the price reversal as
The reversal variable is positive when the sign of the index return before expiration is opposite to that after the expiration. A negative reversal indicates that the market price for the day after the expiration moves in the same direction as that before expiration. If price reversal occurs randomly, probability of the prices moving up or down should be 50 percent each, respectively. A binomial test is used to examine the price reversal effect in the Taiwanese market with the null hypothesis being that price reversals occur randomly.
Abnormal Price Volatility Effect
When index derivatives expire, the strategic actions by speculators or arbitrageurs may drive the spot markets to be more volatile. The standard deviation of the return for the stock index is calculated to estimate the stock market volatility. The windows we set to examine abnormal return volatility are the same as those used in testing abnormal return effect. The null hypothesis is H^sub 0^: S^sub e^; ≤ S^sub c^, where S^sub e^ and S^sub c^ are the standard deviations of the returns of TAIEX on the expiration days and nonexpiration days. A pooled F-test is used to examine whether expirations day returns are associated with higher volatility.
Abnormal Volume Effect
If large underlying stock positions from arbitrageurs or speculators are carried into the expiration day so that the spot market prices are affected, then the trading volume of underlying spot market would be abnormally high on expiration days. To test whether the trading volume difference between expiration days and nonexpiration days exists, we compare the average trading volume on expiration days, V^sub e^ , with the average trading volume of the nonexpiraton days, V^sub c^ , for the experiment period. We use Tuesday, Wednesday entire trading time windows, the first 60-minute and the last 60-minute windows on Wednesday, and the first 15-minute window on Thursday to examine abnormal volume effect. The null hypothesis is H^sub 0^ : V^sub e^ ≤ V^sub c^ and the t-test is used to examine the hypothesis.
Implied Volatility Ssmile Effect
CPA is also used to examine the expiration effect on implied volatility smile. We first observe the closing prices of options with nearest two expiration dates in the same series on each expiration day (Wednesday). The futures price is employed as the proxy of underlying asset of options. In Taiwan, the T-bill is thinly traded in the secondary market. Therefore, the rate of government bond with repurchase agreement is used as the proxy of risk free rate. We calculate implied volatility for each option contract in our sample then average them for each pair of call and put under each level of the exercise price. The control group of nonexpiration day consists of each Wednesday preceding each expiration Wednesday. Finally, we take the methodology adopted by Peña et al. (1999) to estimate the shape of implied volatility smile. The regression equation of implied volatility is:
where σ is the average implied volatility across exercise prices; X is the degree of moneyness, that is the exercise price divided by the futures price. Let K be the exercise price and F is the index future price, then X is the logarithm of K/F; b^sub o^, b^sub 1^, b^sub 2^ are regression coefficients. We regress nearby and second nearby expiration option series on sample days to obtain coefficient b^sub 2^. The b^sub 2^ coefficient is used to describe the shape of implied volatility smile. We average the regression coefficient b^sub 2^ of expiration days and comparison days and the t-test is used to examine whether the average b^sub 2^ coefficients of the expiration days are significantly different from that of the nonexpiration days.
The Determinants of Implied Volatility Smile
From the implied volatility smile literature, the determinants of the volatility smile are still a puzzle. We control the expiration effect determinant of the smile but there might be other factors influencing the smile. To understand the possible determinants of the volatility smile, another four variables are included in the analysis. The first determinant is the annualized standard deviation of the TAIEX which is a measure of uncertainty in the underlying market. The second is the natural log of the trading volume (in thousand NT. dollars). Trading volume presents the level of activity in the underlying market and reflects the information flow therefore it may be related to implied volatility smile. The third is market momentum which reflects the relative level of asset prices. The last one is the TAIEX close to close daily return. The reason why we put the stock return in our analysis is that Black (1976) argued stock return volatility is inversely related to stock returns due to a leverage effect. The regression equation is listed below:
b^sub 2^ 1 is the coefficient determined by equation (12) representing the smile curvature of expiration and nonexpiration days; RS^sub t^ is the TAIEX close to close daily return; VS^sub t^ is the natural log of the trading volume in thousand NT. dollars, MKT^sub t^ is the log of relative market momentum given by ... SIGMA^sub t-1^ is the annualized standard deviation of the TAIEX
The regressions are run for expiration and non-expiration day series to examine whether the determinants of volatility smile on expiration days are the same with on nonexpiration days.
Table 1 reports average stock index returns on expiration day and nonexpiration day group during ten return windows. Hedgers may roll over their positions to next month contracts when expiration day approaches. Thus the abnormal trading activity might take place earlier rather than on expiration day only. From Tuesday window, the day before expiration day, it does not show significant expiration abnormal return effect. For the seven windows on Wednesday, the expiration day average returns are all negative and the control group returns are positive for the first 15, 30, and 60-minute windows. The statistic results show that the first 15, 30, and 60-minute windows on Wednesday all exhibit significant difference at lo% level or stronger. However, the Wednesday all day window and the last 15, 30, 60-minutes windows on Wednesday do not show significant abnormal returns. On Thursday, the settlement day window, the difference between expiration and nonexpiration day average returns is significant at 1% level. From the empirical results, the expiration abnormal return effect exists in the first trading hour on expiration day, Wednesday, and the settlement day, Thursday.
Table 2 presents the pooled F-test results of return volatility of the expiration and nonexpiration windows. The Tuesday window shows the return volatility of the expiration window is more volatile than the nonexpiration window at 10% level. The first 60-minute window on Wednesday also shows significant results. It is worth noticing that the last 15-minute window on Wednesday exhibits the strongest significance at 1% level and the first 15-minute window on Thursday, settlement day, also shows strong significant result at 5% level. These empirical results imply that the arbitragers may try to influence the settlement price not only during the last 15 minutes of trading hour on expiration day but proceed to the first 15 minutes on the settlement day.
The test results of the three measures to capture the possible price-reversal effect are reported in table 3. The first measure compares the index returns during the last sixty minutes before the closing on the expiration day with that closing of the expiration day to the open of the next day, settlement day. There are 59 expiration day samples throughout the sample period, of which 34 exhibit price reversals. The result shows the price reversal effect is significant at 10% level. For the second measure, we extend the second return interval adopted in the first measure to the first fifteen minutes after the opening on the settlement day. The price reversal effect of the second measure exhibits 5% significant level. The result is stronger than the first measure. As the third measure the price reversal is not significant. Recalling the return volatility empirical results, the first 15 minutes after opening of the settlement day experiences strong volatility effect. These evidence shows the expiration effects extend to the settlement day.
Table 4 shows the abnormal volume statistical test results. Stoll and Whaley (1991) and Karolyi (1996) found the abnormal volume occurs on expiration day especially during the last trading hour of the expiration day. However, there are no significant abnormal volume effect occurring on expiration day. The only significant window we examine is the first 15-minute widow on the settlement day. The empirical results provide the evidence that arbitrageurs and hedgers might unwind their position or attempt to influence the settlement price through large order flows during the settlement period.
Implied Volatility Smile Effect
The average implied volatility smile of nearby and second nearby options series on expiration and nonexpiration day is shown in Figure 1. We adopt the moneyness category proposed by Rubinstein (1985), and average the implied volatiles in each category on expiration and nonexpiration day to obtain the average implied volatility smile. From the figure, implied volatility smile seems to be symmetric and the curvature of nearby option series on expiration is much steeper than others. Table 5 reports the t-test statistical results of average smile curvature coefficient b^sub 2^. The table shows that the average b^sub 2^ coefficient of nearby option series on expiration day is 21.23 which is significantly higher than the average b^sub 2^ coefficient, 6.06, on nonexpiration day at a 1% significant level. However, the average b^sub 2^ coefficients of the second nearby option series on expiration day and nonexpiration day do not show significant difference. These results imply that the shape of nearby implied volatility smile grows steeper on the expiration day for nearby options series but not for the second nearby series. The statistical results also present the shapes of implied volatility smile of nearby options series are steeper than the second nearby options series at a 1% significant level on both expiration and nonexpiration days. This finding indicates the nearby option series exhibits stronger expiration effect than the second nearby series. This possible reason is the second nearby options series is still far from expiration so that the turbulent trading activity may not have strong impact on the smile of the second nearby option series.
The Determinants of Implied Volatility Smile
The regression results of the determinants of the implied volatility smile for nearby and second nearby options series on expiration and nonexpiration day are shown in table 6. The degree of significance and explanation power of the regression equation tend to increase with approach of the expiration day. The regression of nearby option series on expiration day is significant at a 1% level and the adjusted R^sup 2^ is 0.235. Nevertheless, the R^sup 2^ of the regression of the second nearby option series on nonexpiration day goes down to -0.018 and the F value shows the regression is not significant. The most dominant results is that the degree of curvature, b^sub 2^, is negatively and significantly related to the historical volatility of the TAIEX. It is also worth noticing the absolute value of the coefficient of SIGMA increases when the expiration day comes near. The results indicate high historical volatility period tend to be associated with lower smile curvature and the nearby options series on expiration day is the most sensitive to the historical volatility. The coefficient of trading volume is negative and significantly different from zero for the nearby option series on the expiration day. The result suggests that the smile pattern across exercise price becomes flatter whenever the trading volume grows.
This study examines the expiration day effects of TAIEX futures and options. The mean return, return volatility, trading volume and volatility smile on the expiration days are compared with those on normal days. Empirical evidences show that abnormal volatility exists on the day before expiration day while the first trading hour on the expiration day exhibits significant, abnormal price return and return volatility. Abnormal volume and abnormal return volatility are found in the first fifteen-minute window on the settlement day while abnormal return and return volatility are found for the entire settlement day. Price reversal is identified in the period from the last trading hour of the expiration day through the first fifteen minutes of trading on the settlement day. The implied volatility smile on the expiration day is significantly different from that on normal days. Moreover, the historical volatility is negatively related to the volatility smile, regardless whether it is the expiration day or not. Our empirical results support the findings that expiration effects are strengthened as more index derivatives are listed on the TAIEX. With the increase of trading volume of index derivatives, the expiration effects do exist on the Taiwan market but tend to shift to the opening period of the settlement day possibly due to its unique settlement mechanism.
Average price settlement procedure is adopted by HKFE and TAIFEX but expiration effects occur in Taiwan market only. The possible reason may be attributed to the length of average price settlement period. HKFE uses average index over the entire expiration day as the settlement price while the settlement period in Taiwan is the first fifteen minutes on the settlement day which is much shorter than that in Hong Kong. If the calculating period of average settlement price is short, arbitrageurs or hedgers still can influence the settlement price through large continuous trading flows. However, the longer the settlement price determination period is, the more difficult the settlement price is manipulated. Thus, longer price determination period might result in lack of arbitrage or hedging activities. The expiration effects are the joint influence of index trading volume and settlement mechanism. It is possible that the short calculating period for settlement price accompanied with huge trading volume of index derivatives cause the expiration effects in Taiwan's market.
This paper also investigates the expiration effect on volatility smiles. It is not surprising the implied volatility smile on the expiration day is different from that on the nonexpiration day. Since the expiration period experiences more turbulent price movements than other times, the underlying return distribution will be leptokurtic in both the right and the left tails of the distribution. Thus log normal risk neural assumption by Black Scholes does not fit the extreme value distribution. Therefore implied volatility smile will reflect the true implicit distribution on the expiration day. The determinants of implied volatility do not show consistent results on the expiration and nonexpiration day. Our findings imply the complexity of implied volatility smile and it is open for further research.
The unique settlement procedure of index derivatives in Taiwan seems to ease expiration effects. Take the abnormal return for example, the most significant difference of index return between the expiration day and the nonexpiration day is 0.53% exhibiting on the settlement day. Although the experimental results show statistically significant level of 1%, after considering the trading cost, the abnormal return may not be economically significant. Therefore it is difficult for arbitrages or hedgers gain abnormal profit around the expiration. Our findings illustrate that the expiration effects, to some extend, might be affected by the settlement procedure. Our results provide insights of the alternative settlement procedure for regulators who concern the abnormal trading activities around expiration days.
Bates, D. S., 2000, "Post-'87 Crash Fears in The S&P 500 Futures options Market, Journal of Econometrics, vol.94, pp. 181-238.
Black, F. and M. Scholes, l973, "The Pricing of Options and Corporate Liabilities", Journal of Political Economy, Vol. 81, pp. 637-659.
Bollen, N. P. B. and R. E. Whaley, 2004, "Does Net Buying Pressure Affect the Shape of Implied Volatility Functions", The Journal of Finance, vol. 59, iss. 2, pp. 711-754.
Chamberlian, T. W., C. S. Cheung, and C. Y. C. Kwan, 1989, "Expiration-Day Effects of Index Futures and Options: Some Canadian Evidence", Financial Analysts Journal, vol.45, Iss. 5, pp. 67-71.
Chan, K. C., L. T. W. Cheng, and P. P. Lung, 2006, "Testing The Net Buying Pressure Hypothesis During The Asian Financial Crisis: Evidence from Hang Seng Index Options", Journal of Financial Research, pp. 43-62.
Chou, H. C., W. C. Chen and D. H. Chen, 2006, "The Expiration Effect of Stock-Index Derivatives" [white square]Emerging Marks Finance and Trade, vol.42, pp.81-102.
Chow, Y. F., H.M. Yung, and H. Zhang, 2003, "Expiration Day Effects: The Case of Hong Kong", Journal of Futures Markets, vol.23, PP-67-86.
Corredor, R., P. Lechon, R. Santamaria, 2001, "Option-Expiration Effects in Small Markets: The Spanish Stock Exchange", Journal of Futures Markets, vol.21, pp.905-928.
Cox, J. and S. A. Ross, 1976 "The Valuation of Options for Alternative Stochastic Processes", Journal of Financial Economics, vol.3, pp. 145 - 166.
Dumas, B. J., and R. E. Whaley, 1998, "Implied Volatility Functions: Empirical Tests", Journal of Finance, vol.53, pp.2059-2106.
Emannuel, D. C, and J. D. Macbethe, 1982, "Further Results on The Constant Elasticity of Variance Call Option Pricing Model", Journal of Financial and Quantitative Analysis, vol.4, PP-533-554-
Hancock, G. D., 1993, "Whatever happened to the triple witching hour?", Financial Analysts Journal, vol.49, iss.3, pp. 66-72.
Hull, J., and A. White, 1987, "The Pricing of Options on Assets with Stochastic Volatilities", Journal of Finance, vol.42, pp.281-300.
Karolyi, A.G., 1996, "Stock Market Volatility around Expiration Days in Japan", Journal of Derivatives,vol.4, iss. 2, pp.23-43.
Masulis, R. W., 1980, "The Effects of Capital Structure Change on Security Prices", Journal of Financial Economics, vol.8, pp. 139-178.
Pena, I., G. Rubio and G. Serna, 1999, "Why Do We Smile? On the Determinants of the Implied Volatility Function" [white square]Journal of Banking & Finance, vol.23, iss. 8, pp. 1151-1179.
Pope, P. F. and P. K. Yadav, 1992, "The Impact of Option Expiration on Underlying Stocks: The UK Evidence", Journal of Business Finance and Accounting, vol.19, pp.329-344.
Rubinstein, M., 1994, "Implied Binomial Trees", Journal of Finance, vol. 49, iss. 3, pp. 771-48.
Schlag, C. 1996,"Expiration Day Effects of Stock Index Derivatives in Germany", European Financial Managernenn vol.1, no.1, pp. 69-95.
Stoll, H. R.and R. E. Whaley, 1986, "Expiration Day Effects of Index Options and Futures", New York University: Monograph Series in Finance and Economics.
Stoll, H. R.and R. E. Whaley, 1987, "Program Trading and Expiration Day Effects", Financial Analysts Journal, vol.43, iss. 2, pp. 16-28.
Stoll, H. R., and R. F. Whaley, 1991, "Expiration-Day Effects: What Has Changed?", Financial Analysts Journal, vol. 47, pp.58-72.
Stoll, H. R. and R. E. Whaley, 1997, "Expiration-Day Effects of the All Ordinaries Share Price Index Futures [white square] Empirical Evidence and Alterative Settlement Procedures", Australian Journal of Management, vol.22, .pp. 139-174.
Swidler, S., L. Schwartz and Kristiansen, 1994, "Option Expiration Day Effects in Small Market; Evidence from the Oslo Stock Exchange", Journal of Financial Engineering, vol, 3, pp.177-195.
Shiaw-En Ju - National Central University in Taiwan, R.O.C.
Keng-Hsin Lo - National Central University, Taiwan, R.O.C.
Kehluh Wang - National Chiao Tung University, Taiwan, R.O.C.
Shiaw-En Ju is a doctoral student from Taiwan. He earned his first masters degree in Hydraulic Engineering at Taiwan University in 1994. He subsequently turned his research interest to business. In 2001 , he graduated from Cardiff Business School in the United Kingdom with an MBA degree. Currently, he is a PhD student at the National Central University in Taiwan. His research interests entail corporate finance, financial engineering and market microstructure.
Keng-Hsin Lo is an associate professor of National Central University. He obtained his Ph.D. degree from National Chengchi University in 1990. His research interests are financial management and strategy management
Kehluh Wang is an associate professor of National Chiao Tung University. He received Ph.D. degree from Northwestern University in 1989. His research interests are financial engineering and corporate finance…
Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Article title: Expiration Day Effects: Empirical Evidence from Taiwan. Contributors: Ju, Shiaw-En - Author, Lo, Keng-Hsin - Author, Wang, Kehluh - Author. Journal title: Journal of Global Business Issues. Publication date: Spring 2008. Page number: 51+. © Journal of Global Business Issues Summer 2008. Provided by ProQuest LLC. All Rights Reserved.
This material is protected by copyright and, with the exception of fair use, may not be further copied, distributed or transmitted in any form or by any means.