You are on page 1of 29

Asia-Pacific Finan Markets (2013) 20:83–111

DOI 10.1007/s10690-012-9160-1

Forecasting Intraday Volatility and Value-at-Risk


with High-Frequency Data

Mike K. P. So · Rui Xu

Published online: 9 November 2012


© Springer Japan 2012

Abstract In this paper, we develop modeling tools to forecast Value-at-Risk and


volatility with investment horizons of less than one day. We quantify the market risk
based on the study at a 30-min time horizon using modified GARCH models. The
evaluation of intraday market risk can be useful to market participants (day traders
and market makers) involved in frequent trading. As expected, the volatility features
a significant intraday seasonality, which motivates us to include the intraday sea-
sonal indexes in the GARCH models. We also incorporate realized variance (RV) and
time-varying degrees of freedom in the GARCH models to capture more intraday
information on the volatile market. The intrinsic tail risk index is introduced to assist
with understanding the inherent risk level in each trading time interval. The proposed
models are evaluated based on their forecasting performance of one-period-ahead vol-
atility and Intraday Value-at-Risk (IVaR) with application to the 30 constituent stocks.
We find that models with seasonal indexes generally outperform those without; RV
can improve the out-of-sample forecasts of IVaR; student GARCH models with time-
varying degrees of freedom perform best at 0.5 and 1 % IVaR, while normal GARCH
models excel for 2.5 and 5 % IVaR. The results show that RV and seasonal indexes
are useful to forecasting intraday volatility and Intraday VaR.

Keywords GARCH · Intraday market risk · Intrinsic tail risk index ·


Realized volatility · Risk management · Seasonality · Value at Risk

M. K. P. So (B)
Department of Information Systems, Business Statistics and Operations Management,
School of Business and Management, The Hong Kong University of Science
and Technology, Clear Water Bay, Hong Kong
e-mail: immkpso@ust.hk

R. Xu
Department of Economics, Stanford University, Stanford, CA, USA

123
84 M. K. P. So, R. Xu

1 Introduction

High frequency data has become readily available to all as a result of recent
advances in trading technology. Grasping the opportunities that this presents, a new
group of active market participants, such as high-frequency traders, has emerged
in the major financial markets. These practitioners are characterized by their very
short investment horizons, and their (heavy) reliance on tools for market risk mea-
surement. As risk must be estimated in intraday intervals, traditional risk mea-
sures, such as Value at Risk (VaR) and return volatility, have to be extended to
cater for shorter than one-day time intervals. As demonstrated during the global
financial crisis in late 2008, even the price of blue-chip stocks can fall sub-
stantially within a day. This kind of dramatic intraday price movements in the
equity market can cause huge losses for many active traders and market makers.1
Here, we investigate the intraday forecasts of volatility and VaR using tick-by-
tick transactions with an aim to improve the tools for measuring the market risk.
The use of high-frequency data, defined as intraday market information includ-
ing transaction prices, bid-ask prices and intraday trading volume, etc, has been
studied for more than ten years. Early researches have concentrated on exchange
rates, e.g. Zhou (1996), Taylor and Xu (1997) and Beltratti and Morana (1999).
More recent works with a focus on equity data include Andersen et al. (2001a),
Giot and Laurent (2004) and Fuertes et al. (2009). The above-mentioned stud-
ies have centered on the calculation of daily risk measures, whereas this paper
develops methodologies for forecasting risk measures with a horizon of less than
a day.
The major contributions of our research rest on four aspects. First, we examine the
distribution of intraday equity returns and realized volatilities with high-frequency
data of Hang Seng Index’s (HSI) stocks. We choose HSI because it is represen-
tative of the statistical properties of stock intraday returns in Hong Kong. To the
best of our knowledge, this is also the first comprehensive study of intraday stock
return distribution in Asia. Our research on the 30-min return distributions presents a
good comparative case to similar studies on the daily return distributions of the S&P
500 index (Andersen and Bollerslev 1998), the Dow Jones Industrial Average stocks
(Andersen et al. 2001a), the NYSE stocks (Fuertes et al. 2009) and the Spanish Stock
Exchange (Coroneo and Veredas 2006). From the summary statistics, we find that
the distribution of intraday equity returns and realized volatilities of HSI constituent
stocks share similar properties with the daily equity return and realized volatility dis-
tribution presented in Andersen et al. (2001a), but the intraday distributions exhibit
clear seasonality which should be accommodated in forecasting exercises. Specif-
ically, we identify a significant W-shaped intraday volatility pattern, which serves
as the major motivation for us to incorporate seasonal indexes in intraday volatility
modeling.
Second, we propose simultaneous modeling of seasonality in volatility and tail
risk under GARCH. Our approach explains heterogeneous volatility levels and

1 For example, CITIC Pacific, which is a blue-chip constituent of thein Hong Kong Stock Exchange,
dropped by as much as 28.5 % on October 27th, 2008 to close at 3.66 HKD.

123
Forecasting Intraday Volatility and Value-at-Risk 85

different tail characteristics in different intraday time periods. In our volatility


forecasting models, intraday seasonal indexes are incorporated to account for the sig-
nificant seasonal patterns demonstrated in the descriptive statistics of 30-min returns.
In past literature, the intraday seasonal patterns were usually filtered before modeling
volatility as in Giot (2005), Coroneo and Veredas (2006) and Dionne et al. (2009);
while in our study, we include seasonal indexes in the conditional variance prediction
model, together with time-dependent tail characteristic parameters, to improve out-of-
sample forecasting performance. As it turns out, models with seasonal indexes out-
perform models without in terms of the ability to predict intraday volatility and VaR.
In addition, we introduce the intrinsic tail risk index to capture the part of VaR
due to the seasonality and the inherent risk in each time interval. This index can
help investors to pre-assess the intraday trading risk with the most recent market
condition.
Third, we incorporate 30-min realized variance (RV) into the volatility forecasting
model. RV, a sum of squared returns with moderate frequency data, is a natural mea-
sure of variability of the price path (see Andersen et al. 2001b; Barndorff-Nielsen and
Shephard 2002; Zhang et al. 2005; Hansen and Lunde 2006). The nonparametric nature
of RV and the simplicity of its calculation have made it popular among practitioners.
It has been used for asset allocation (Fleming et al. 2003), the forecasting of Value at
Risk (Giot and Laurent 2004), evaluation of volatility forecasting models (Andersen
and Bollerslev 1998) and the building of stochastic volatility models (Takahashi et al.
2009), among other purposes. Unlike past studies where daily RV is used as a proxy
for daily latent return volatility (Martens 2001, 2002; Koopman et al. 2005; Brown-
lees and Gallo 2010), we incorporate intraday RV in the conditional volatility model
to capture the most recent information and help provide better forecasts of intraday
volatility. RV is considered in GARCH models in two ways: one is to replace 30-min
squared return with RV, the other is to add an additional term of RV into GARCH
models.
Finally, we derive a forecasting formula for |Rt+1 |c under conditional t distri-
butions to assess volatility predictive performance. Based on the prediction perfor-
mances of powered absolute returns, our novel use of RV significantly improves the
volatility forecasting performance for a 30-min investment horizon. We also use the
conditional VaR framework to study the performance of VaR prediction by modified
GARCH-t models with time-varying degrees of freedom. Each model is evaluated
and scored according to its ranking in forecasting accuracy. The overall performance
of the models, obtained by averaging the ranking scores, shows that seasonal indexes
and time-varying degrees of freedom can improve out-of-sample forecasts of intraday
VaR.
The remainder of this paper is organized as follows. Section 2 describes the data
sets used in the study and the statistical properties of 30-min stock return and realized
volatility. In Sect. 3, we present the construction of three groups of modified GARCH
models to forecast intraday volatility. The intrinsic tail risk index is also introduced.
In Sect. 4, volatility forecasting and VaR forecasting performances of all models are
evaluated with application to the empirical data of 30 HSI constituent stocks. Outper-
forming models are identified for volatility and intraday VaR forecasting. Section 5
concludes the paper.

123
86 M. K. P. So, R. Xu

2 Data and Descriptive Statistics

2.1 Data Description and Cleaning

Our data set, obtained from the History Tool of Bloomberg Station, contains tick-by-
tick transaction information (time, price and volume) of 30 constituent stocks of the
HSI. The 30 stocks are selected as follows: we start by including all 42 constituent
stocks of the HSI; then we eliminate those that are not the major constituents of the
Hang Seng HK LargeCap Index, the HK MidCap Index or the Mainland Compos-
ite Index. Finally, stocks with erroneous data are excluded. The sample time period
starts on March 25th 2008 and ends on May 31st 2009, which covers the most vola-
tile period of the subprime financial crisis. The 30 stocks are Cheung Kong Holdings
Ltd (CK), CLP Holdings Ltd (CLP), Wharf Holdings Ltd (Wharf), HSBC Holdings
plc (HSBC), Hong Kong Electric Holdings Ltd (HK Electric), Hang Seng Bank Ltd
(HS Bank), Henderson Land Development Co. Ltd (Henderson), Hutchison Wham-
poa Ltd (Hutchison), Sun Hung Kai Properties Ltd (SHK), New World Development
(New World), Swire Pacific Ltd ‘A’ (Swire A), MTR Corporation Ltd (MTR), Sino
Land Co. Ltd (Sino Land), Hang Lung Properties Ltd (Hang Lung), CITIC Pacific
Ltd (CITIC Pacific), Cathay Pacific Airways Ltd (Cathay Pacific), Sinopec Corp (Sin-
opec), HKEx Limited (HKEx), Li & Fung Ltd (Li & Fung), China Unicom (China
Unicom), PetroChina Co. Ltd (PetroChina), CNOOC Ltd (CNOOC), China Construc-
tion Bank (CCB), China Mobile Ltd (China Mobile), Industrial and Commercial Bank
of China (ICBC), Foxconn International Holdings Ltd (FIH), Ping An Insurance (Ping
An), Aluminum Corporation of China Ltd (CHALCO), China Life (China Life) and
Bank of China Ltd (BOC).
To help with better understanding the intraday data of the 30 stocks, we briefly
describe some distinguished features of the Hong Kong Stock Exchange (HKEx). As
Asia’s third largest stock exchange in terms of market capitalization, the HKEx has
more than 1,145 listed companies with a combined market capitalization of 17,769,271
as of December 2009. The most widely quoted indicator of the HKEx’s performance
is the Hang Seng Index (HSI). The 42 constituent stocks of the HSI are classified into
one of four sub-indexes: Finance, Utilities, Properties, and Commerce and Industry.
The 30 stocks studied in this paper cover all four sub-indexes; 12 of them are also
composites of the Hang Seng HK LargeCap Index (HSHKLI), 6 are composites of the
Hang Seng HK MidCap Index (HSHKMI) and 12 are composites of the Hang Seng
Mainland Composite Index (HSMLCI). The trading hours of HKEx are different from
other major stock markets. A normal trading day consists of four sessions: a pre-open-
ing auction session from 9:30 a.m. to 9:50 a.m.; a morning continuous trading session
from 10:00 a.m. to 12:30 p.m.; an afternoon continuous trading session from 14:30 to
16:00; and a closing auction session from 16:00 to 16:10.2 The high-frequency trans-
action data we use include information for all trading sessions of each day. Before

2 The closing auction session was implemented in May 2008, but it was removed in March 2009 due to
significant fluctuations in the closing prices of stocks. News releases of the implementation and suspension
can be found at the following links: http://www.hkex.com.hk/news/hkexnews/0805192news.htm and http://
www.hkex.com.hk/news/hkexnews/090320news.htm.

123
Forecasting Intraday Volatility and Value-at-Risk 87

computing the 30-min equity return and the realized volatility, we take two steps to
clean the data sets. First, trades in the closing auction session are ignored, because the
session has been removed in March 2009. Second, transactions with odd lot, whose
trading volume is not exact multiples of the lot size, are removed because they usually
have big jumps compared to their neighbor transaction prices. After data cleaning, we
calculate the 30-min return and the realized volatility of the 30 constituent stocks.
To enable the study of intraday risk in a 30-min horizon, we define Pt,i , where
i = 1, . . ., 8, as the closing price in the ith interval on day t, which is the last transac-
tion price in each of the eight half-hour intervals [10:00, 10:30), . . ., [12:00, 12:30),
[14:30, 15:00), . . ., [15:30, 16:00). Then, the logarithm return of an asset in the ith
30-min intraday interval on day t is taken to be rt,i = (log Pt,i − log Pt,i−1 )× 100 %.
In our study, t refers to the trading days between March 25th 2008 and May 31st 2009
and i takes on the value of 1–8. In particular, we define Pt,0 as the last transaction
price in the pre-opening auction session in order to get rt,1 . As for the calculation of
realized volatility, we follow the simple practice of summing up the frequently sam-
pled squared returns. This approach is justified under the assumption of a continuous
stochastic model. However, in the presence of microstructure noise, such as bid-ask
bounce, asynchronous trading, infrequent trading, and price discreteness, the more
frequently stock returns are sampled, the more volatility will be overestimated. To
determine the optimal sampling frequency, we follow the argument of Andersen and
Bollerslev (1998) and use the calendar time sampling scheme to obtain the RV by
summing up the six 5-min squared returns in the 30-min intraday interval. We define
the 30-min RV r vt,i 2 as follows.


6
 2
r vt,i
2
= log Pt,i,n − log Pt,i,n−1 , i = 1, . . . , 8,
n=1

where Pt,i,n is the price just before the end of the nth 5-min interval in the ith 30-min
interval on day t.

2.2 Unconditional Distributions

After calculating the 30-min return and realized volatility series for 30 stocks, we dis-
play their unconditional distributions in Tables 1 and 2 in a way similar to Andersen
et al. (2001a), yet for a 30-min intraday horizon. Daily return and realized volatility
distributions are shown in Tables 3 and 4 for comparison. The left panel of Table 1
refers to the 30-min unconditional distributions of the 30 mean returns, standard devi-
ations, skewnesses and excess kurtosis. Most mean returns in our data set are negative
due possibly to the effect of the financial tsunami since late 2007. The majority of
the 30-min return skewnesses are positive. Compared with the daily return distribu-
tion in Table 3, the skewnesses of the 30-min data are more dispersed, ranging from
−1.150 to 4.470. The excess kurtosis of rt,i indicates more severe fat-tailed behavior
than the daily data. The median excess kurtosis is 8.265, which is greater than 4.285
of the daily return, suggesting more severe fat-tailed behavior for intraday returns.

123
88 M. K. P. So, R. Xu

Table 1 Unconditional 30-min return distributions (%)

Stock rt,i rt,i /r vt,i

Mean SD Skewness Kurtosis Mean SD Skewness Kurtosis

Min −0.085 0.654 −1.150 5.420 −0.081 0.798 −0.090 −0.780


0.100 −0.047 0.807 −0.375 6.327 −0.049 0.878 −0.023 −0.701
0.250 −0.027 0.902 −0.205 6.653 −0.038 0.884 −0.008 −0.665
0.500 −0.008 1.051 0.105 8.265 −0.017 0.901 0.020 −0.620
0.750 −0.002 1.243 0.383 15.245 −0.005 0.914 0.030 −0.573
0.900 0.008 1.385 0.500 20.449 0.008 0.939 0.041 −0.468
Max 0.026 1.931 4.470 106.740 0.033 1.179 1.600 26.240
Mean −0.015 1.106 0.198 14.183 −0.020 0.907 0.060 0.286
The table summarizes the 30-min return distributions for the 30 HSI constituent stocks, rt,i . The sample
covers the period from March 25, 2008 through May 31, 2009 for a total of 2,300 observations. The 30-min
realized volatilities, r vt,i , are calculated from 5-min intraday returns, as detailed in the main text

Table 2 Unconditional 30-min realized volatility distributions (%)

Stock r vt,i log(r vt,i )

Mean SD Skewness Kurtosis Mean SD Skewness Kurtosis

Min 0.519 0.404 2.430 9.580 −0.887 0.527 −2.830 −0.180


0.100 0.690 0.518 2.735 12.096 −0.620 0.570 −0.610 0.160
0.250 0.820 0.606 2.885 14.995 −0.447 0.618 −0.498 0.783
0.500 0.929 0.705 3.135 18.440 −0.256 0.670 −0.160 1.925
0.750 1.075 0.829 3.425 23.538 −0.165 0.717 −0.020 3.258
0.900 1.177 0.973 4.528 50.793 −0.055 0.763 0.110 4.289
Max 1.613 1.324 9.830 220.510 0.215 0.802 0.360 19.340
Mean 0.949 0.733 3.662 34.190 −0.302 0.668 −0.295 2.579
The table summarizes the distributions of the 30-min realized volatilities for the 30 HSI constituent stocks,
r vt,i . The realized volatilities, r vt,i and logarithmic standard deviations, log(r vt,i ), are calculated from
5-min intraday returns, as detailed in the main text

In the panel on the right, the mean standardized returns, rt,i /r vt,i , are close to zero.
The standard deviations of rt,i /r vt,i are close to one as in the daily data. After stan-
dardization, the distribution is closer to but still not exactly Gaussian according to the
Anderson-Darling test (the histograms for HSBC before and after standardization are
shown in Fig. 1a, b). It is in fact quite different from the daily standardized return,
which is approximately normal as shown in Fig. 1c. The major difference, again, lies
in the excess kurtosis of rt,i /r vt,i , which is smaller and mostly negative for 30-min
standardized returns.
In Table 2, we present unconditional distributions of the summary statistics of the
30-min realized volatility. Despite having smaller mean r vt,i than the daily data, the
standard deviations of the 30-min log(r vt,i ) are generally larger than that of the daily
data. The results suggest higher volatility in 30-min realized volatility than in daily

123
Forecasting Intraday Volatility and Value-at-Risk 89

Table 3 Unconditional daily return distributions (%)

Stock rt,i rt,i /r vt,i

Mean SD Skewness Kurtosis Mean SD Skewness Kurtosis

Min −0.425 1.855 −1.590 1.280 −0.135 0.768 −0.430 −0.620


0.1 −0.263 2.207 −1.237 1.749 −0.088 0.832 −0.228 −0.442
0.25 −0.142 2.677 −0.260 2.293 −0.060 0.863 0.015 −0.410
0.5 −0.037 3.036 0.105 4.285 −0.035 0.890 0.075 −0.100
0.75 0.043 3.477 0.408 6.765 −0.002 0.983 0.153 0.253
0.9 0.089 3.942 0.603 12.553 0.018 1.006 0.270 0.634
Max 0.255 6.016 0.930 17.660 0.082 1.021 0.390 3.410
Mean −0.056 3.145 −0.076 5.781 −0.033 0.904 0.063 0.102

Table 4 Unconditional daily realized volatility distributions (%)

Stock r vt,i log(r vt,i )

Mean SD Skewness Kurtosis Mean SD Skewness Kurtosis

Min 1.642 0.963 1.390 2.780 0.351 0.338 −0.170 −0.410


0.1 2.246 1.106 1.811 4.895 0.703 0.372 0.228 0.035
0.25 2.625 1.305 2.028 5.793 0.847 0.403 0.305 0.200
0.5 2.867 1.566 2.410 8.415 0.959 0.434 0.475 0.550
0.75 3.424 1.772 2.705 12.075 1.124 0.478 0.675 1.225
0.9 3.734 2.236 4.532 37.438 1.232 0.531 0.800 1.709
Max 5.132 4.522 14.570 234.440 1.514 0.579 1.020 3.890
Mean 3.024 1.679 2.981 19.882 0.970 0.442 0.491 0.799

realized volatility after eliminating the scale effect. The third and fourth columns
demonstrate that the distributions of the 30-min realized volatilities are extremely
right-skewed and leptokurtic. After logarithm transformation, interestingly, the median
value of the sample skewness across all of the 30 stocks is reduced to -0.160, com-
pared to 3.135 before transformation. In the panel on the right, the skewness and
excess kurtosis of log (r vt,i ) indicate that log(r vt,i ) is roughly symmetric and has low
leptokurtosis, though all but one of the stocks’ excess kurtosis exceeds the normal
value of zero. As shown in Fig. 2b, the Gaussian assumption made on the distribution
is plausible. This evidence is consistent with the findings in Andersen et al. (2001a,b).

2.3 Return and Realized Volatility Distributions in Eight 30-min Intervals

After studying the unconditional distributions, we conduct a separate analysis for


the eight 30-min intervals, [10:00, 10:30), . . ., [12:00, 12:30), [14:30, 15:00), . . .,
[15:30, 16:00). Table 5 displays the distribution of returns, rt,i , and standardized
returns, rt,i /r vt,i , for i = 1-8. First, by comparing the median statistics of the return
distributions in the left panel, we can identify significant W-shaped intraday seasonal

123
90 M. K. P. So, R. Xu

(a) Histogram of 30-min return - HSBC


Normal
700

600

Frequency 500

400

300

200

100

0
-5.25 -3.50 -1.75 0.00 1.75 3.50 5.25
30- min return of HSBC

(b) Histogram of standardized 30-min return


Normal
250

200
Frequency

150

100

50

0
-1.8 -1.2 -0.6 0.0 0.6 1.2 1.8 2.4
Standardized return

(c) Histogram of standardized daily return - HSBC


Normal
40

30
Frequency

20

10

0
-2.25 -1.50 -0.75 0.00 0.75 1.50 2.25
standardized return

Fig. 1 Distribution of 30-min return for HSBC. a Unconditional 30-min raw return distribution,
b standardized 30-min return distribution, c standardized daily return distribution

123
Forecasting Intraday Volatility and Value-at-Risk 91

(a)

(b)

Fig. 2 Distribution of 30-min realized volatility for HSBC. a Unconditional 30-min realized volatility
distribution, b logarithm realized volatility distribution

patterns: the standard deviation of rt,i shows that stock prices are most volatile in the
first 30 min of trading at 10:00 a.m.–10:30 a.m., followed by the 30 min of trading at
14:30–15:00 after the lunch break, and 15:30–16:00 before the market closes. This
seasonal pattern of the volatility is reasonable, as news released overnight or during
the lunch break may lead to significant price movement right after the market reopens.
At 15:30–16:00, day-traders would liquidate any open positions at closing, in order to
pre-empt any adverse overnight moves resulting in large gap openings of the next trad-
ing day. Also, the mean kurtosis of return peaks at 12:00–12:30 indicates extremely
large fluctuations in price before the lunch break. Moreover, we identify three inter-
vals with interesting uniform positive/negative returns: at 11:30–12:00, all but two
(CITIC Pacific and FIH, which are also the most volatile stocks amongst all) stocks
have positive mean returns; at 14:30–15:00 right after lunch break, all 30 stocks have
negative mean returns; at 15:30–16:00 before the market closes, all but two (Wharf
and CITIC Pacific) stocks have positive mean returns. The first of these intervals is
especially worth attention, because positive return is not common during the financial

123
92 M. K. P. So, R. Xu

Table 5 Median of the four statistics (mean, standard deviation, skewness and excess kurtosis) of 30-min
return in eight intraday intervals

Time interval rt,i rt,i /r vt,i

Mean SD Skewness Kurtosis Mean SD Skewness Kurtosis

10:00–10:30 −0.110 1.584 0.120 2.840 −0.062 0.948 0.060 −0.830


10:30–11:00 0.003 0.904 −0.170 6.020 −0.026 0.896 −0.025 −0.535
11:00–11:30 −0.050 0.765 −0.210 4.230 −0.083 0.887 0.030 −0.560
11:30–12:00 0.046 0.780 0.690 6.780 0.040 0.871 −0.060 −0.495
12:00–12:30 0.019 0.848 1.820 12.480 −0.018 0.884 0.020 −0.590
14:30–15:00 −0.091 1.258 −0.050 5.700 −0.062 0.928 0.065 −0.740
15:00–15:30 0.003 0.995 0.020 5.740 −0.013 0.919 0.025 −0.585
15:30–16:00 0.069 1.041 −0.050 4.690 0.076 0.874 −0.120 −0.500
Bold values refer to the W-shaped seasonal pattern
Italic values refer to the extreme fluctuation in price before lunch break

crisis period and traders may make use of this pattern in their day trading algorithm to
make profits. The negative return interval occurs right after the lunch break, indicating
that the significant price movements in this interval are mainly downside. The positive
mean return before the closing probably indicates that buying between 15:00–15:30
and liquidating the position between 15:30–16:00 may result in profit in the long
run. After standardizing the returns, it is striking to see that similar seasonal patterns
remain: the standard deviation of standardized return still peaks at the market opening
in the morning and the market reopening right after the lunch break. This indicates
that seasonal patterns of intraday return cannot be fully explained by the seasonality
in realized volatility, thus we propose to incorporate seasonal indexes in our con-
ditional volatility modeling. In addition, the kurtosis of standardized returns ranges
from −0.830 at 10:00–10:30 to −0.495 at 11:30–12:00. This motivates us to consider
seasonality in the tail heaviness for the intraday volatility and VaR estimation.
The findings in Table 6 also reflect strong seasonality in intraday volatility. In
accordance with the seasonal patterns of the intraday return series, realized volatility
demonstrates similar seasonal patterns, with the first peak at 10:00–10:30, and the
second peak at 14:30–15:00 right after the lunch break, as shown in column 1 of
Table 6. Fat-tailed behavior is evident in r vt,i for all i = 1–8, with the largest kurtosis
occurring at 14:30–15:00. After logarithm transformation, it is evident in the right
part of Table 6 that the seasonal patterns are retained (as logarithm is a monotone
transformation) and the distributions of log(r vt,i ) are much closer to Gaussian.
In summary, a significant seasonal pattern is present in the level of volatility and
the kurtosis of 30-min intraday returns. We should take it into account when construct-
ing volatility models. Seasonal indexes and time-varying degrees of freedom will be
incorporated in the volatility forecasting models to be introduced in Sect. 3.

3 GARCH Modeling with Seasonality in Volatility and the Tail Risk

In the past, various studies (Giot 2005; Hansen and Lunde 2005) have been conducted
to compare the performance of different GARCH-type models in forecasting daily

123
Forecasting Intraday Volatility and Value-at-Risk 93

Table 6 Median of the four statistics (mean, standard deviation, skewness and excess kurtosis) of 30-min
realized volatility in eight intraday intervals

Time interval r vt,i log(r vt,i )

Mean SD Skewness Kurtosis Mean SD Skewness Kurtosis

10:00–10:30 1.448 1.013 2.350 8.525 0.185 0.583 0.070 0.215


10:30–11:00 0.873 0.528 2.035 6.190 −0.291 0.572 −0.030 0.360
11:00–11:30 0.774 0.465 1.960 6.045 −0.411 0.589 −0.185 0.215
11:30–12:00 0.735 0.460 2.060 6.440 −0.452 0.645 −0.705 3.020
12:00–12:30 0.740 0.534 2.500 9.615 −0.460 0.650 −0.050 0.650
14:30–15:00 1.135 0.846 2.930 12.600 −0.064 0.648 0.160 0.495
15:00–15:30 0.859 0.591 2.765 12.280 −0.331 0.646 −0.125 0.830
15:30–16:00 0.914 0.608 2.755 11.275 −0.243 0.609 −0.020 0.260
Bold values refer to the strong seasonality in intraday volatility

volatility. The ordinary GARCH(1,1) model was shown to provide satisfactory results.
Therefore, we use the framework of the GARCH(1,1) model in all three groups of
models. One of our major modifications is to incorporate RV in GARCH(1,1) models,
2 with r v 2 , or by adding r v 2 as an extra term. As mentioned at
either by replacing rt,i t,i t,i
the end of Sect. 2, we also consider seasonal patterns in one group of models, and both
seasonal patterns and changing degrees of freedom in the third group. The detailed
model structures of different groups are introduced below.

3.1 Modified GARCH Models

The first group of modified GARCH(1,1) models focuses on the use of RV as one of
the predictors of the conditional variance of the 30-min return rt,i . In all the modified
GARCH models we develop in this paper, we assume that rt,i is generated by

rt,i = σt,i × εt,i , i = 1, 2, . . . , 8. (1)

The major differences among various models lie in the formulation of σt,i , the con-
ditional variance of rt,i given the previous information up to the (i −1)th interval on
day t, t,i−1 , and in the specification of the distribution of εt,i . The first group of six
models is defined as follows.
σt,i
2
= α0 + α1 rt,i−1
2
+ β1 σt,i−1
2
, εt ∼ N (0, 1), (Model 1Gn )

σt,i
2
= α0 + α1r vt,i−1
2
+ β1 σt,i−1
2
, εt ∼ N (0, 1), (Model 1Rn )

σt,i
2
= α0 + α1rt,i−1
2
+ β1 σt,i−1
2
+ γ1r vt,i−1
2
, εt ∼ N (0, 1), (Model GRn )

σt,i
2
= α0 + α1rt,i−1
2
+ β1 σt,i−1
2
, εt ∼ T (0, 1, ν), (Model 1Gt )

123
94 M. K. P. So, R. Xu

σt,i
2
= α0 + α1r vt,i−1
2
+ β1 σt,i−1
2
, εt ∼ T (0, 1, ν), (Model 1Rt )

σt,i
2
= α0 + α1rt,i−1
2
+ β1 σt,i−1
2
+ γ1r vt,i−1
2
, εt ∼ T (0, 1, ν),
(Model 1GRt )
where r vt,i
2 is the RV defined in Sect. 2; T (0, 1, ν) refers to a t distribution with mean

0, variance 1 and a degree of freedom ν which is assumed to be greater than 4. As


usual, we also restrict α0 > 0 and α1 + β1 < 1. From Eq. (1), we model the returns of
the first to the eighth 30-min interval using previous returns. In particular, we estimate
the first 30-min return rt,1 using rt−1,8 and r vt−1,8 of the previous trading day. Of the
above six GARCH specifications, Models 1Gn and 1Gt are the basic GARCH-normal
and student GARCH models. In Models 1Rn and 1Rt , we replace rt,i−1 2 with r vt,i−1
2 ,
the RV, which is thought to be a good volatility proxy, especially for intraday market
movement. Models 1GRn and 1GRt are formed by adding r vt,i−1 02 as an exogenous
variable to Models 1Gn and 1Gt , respectively. In essence, the first group of modifi-
cations aims to investigate the power of RV for predicting future volatility and VaR.
The model parameters, α0 , α1 , β1 , γ1 and ν are estimated using the Maximum Like-
lihood (MLE) approach. The log-likelihood function for models with normal-error
distribution and models with t-error distribution are as follows:

 

N 
8
1
2
rt,i
log L = − log(2π ) − log(σt ) − 2
, for εt,i ∼ N (0, 1)
2 2σt,i
t=1 i=1
N  8  
ν 1
1 ν+1
log L = − log − log − log(π )
2 2 2 2
t=1 i=1
1
− log(ν − 2) − log(σt,i )
2 
2
rt,i
1
− (ν + 1) log 1 + 2 , for εt,i ∼ T (0, 1, ν) (2)
2 σt,i (ν − 2)

where the initial value of σ12 is set to be the variance of rt ; (.)is the gamma function;
N is the total number of 30-min returns for each stock from March 25th 2008 to May
31st 2009.
In the second group of modified GARCH models, we introduce multiplicative sea-
sonal indexes to reflect the characteristics of the intraday variation due to the 30-min
segmentation of the 4-h trading period and any possible intraday periodicity. Similar
to the model structure of models in group 1, a multiplicative seasonal index S(i) is
added to the models in group 2 in the volatility equation of σt,i 2 . Specifically,

σt,i
2
= S(i) × τt,i
2
, (3)

123
Forecasting Intraday Volatility and Value-at-Risk 95

where τt,i
2 is called the adjusted conditional variance and is taken as the right-hand side

of any of Models 1Gn − 1GRt and S(i)(i = 1, 2, . . ., 8) are positive


8 seasonal indexes
for various 30-min changes. To identify the model in (3), we set i=1 S(i) = 1. The
log-likelihood functions are the same as those in the models in group one. We label
the six models with seasonal indexes as Model 2Gn , 2Rn , 2GRn , 2Gt , 2Rt and 2GRt .
For instance, Model 2Gn is defined as

τt,i
2
= α0 + α1rt,i−1
2
+ β1 σt,i−1
2
,


σt,i
2
= S(i) × α0 + α1rt,i−1
2
+ β1 σt,i−1
2
, εt,i ∼ N (0, 1), (Model 2Gn )

and Model 2GRt as

τt,i
2
= α0 + α1rt,i−1
2
+ β1 σt,i−1
2
+ γ1r vt,i−1
2
,


σt,i = S(i) × α0 + α1rt,i−1 + β1 σt,i−1 + γ1r vt,i−1
2 2 2 2
, εt ∼ T (0, 1, ν).
(Model 2GRt )

The novelty of the group 2 models defined in (3) is that the conditional variance
σt,i
2 is decomposed into two parts; one is the adjusted conditional variance τ 2 and the
t,i
other is the seasonal index.
√ The decomposition implies that the conditional√ variance of
adjusted return, rt,i / S(i), is equal to τt,i
2 . In other words, var(r / S(i)|
t,i t,i−1 ) =
τt,i
2 . From Eq. (3) and the above constraint, the geometric mean of the σ 2 , . . ., σ 2 is
t,1 t,8
identical to the geometric mean of τt,1
2 , . . ., τ 2 , which is independent of S(i). We can
t,8
interpret τt,i
2 as the explained return variation attributed to previous return movement

and S(i) as an intrinsic scale factor independent of the market movement.


The third group of modified GARCH models combines seasonal indexes and time-
varying degrees of freedom together. As documented in Sect. 2, there is evidence of
intraday variation in the kurtosis or more generally of a tail risk, which can be quan-
titatively described by the degrees of freedom of the t distribution characterizing the
error εt,i . We define Models 3Gt , 3Rt and 3GRt , from Models 2Gt , 2Rt and 2GRt to
incorporate both seasonal indexes and to allow ν, the degrees of freedom, to vary in
different intraday intervals. We also relax the assumption for ν by reducing its lower
bound from 4 to 2. For instance, Model 3Gt is defined as

τt,i
2
= α0 + α1rt,i−1
2
+ β1 σt,i−1
2
,


σt,i
2
= S(i) × α0 + α1rt,i−1
2
+ β1 σt,i−1
2
, εt ∼ T (0, 1, νi ). (Model 3Gt )

The major advantage of the group 3 models is that we can have flexible tail struc-
tures in 30-min returns, which enables financial analysts to monitor tail risk on an
intraday basis. On top of the GARCH parameters, we have eight pairs of S(i) and νi
to be estimated by maximizing the log-likelihood function below

123
96 M. K. P. So, R. Xu

N  8  
ν 1
1 νi + 1 i 1
log L = − log − log − log(π ) − log(νi − 2)
2 2 2 2 2
t=1 i=1
2

1 rt,i
− log(σt,i ) − (νi + 1) log 1 + 2 ,
2 σt,i (νi − 2)
where νi is the time-varying degrees of freedom in the ith intraday interval.

3.2 Intraday Seasonality Results

We compute the parameters of the 15 models in the three groups introduced in Sect. 3.1
using the training data set from March 25th 2008 to October 31st 2008. The estimated
parameters for the 15 models are summarized in Table 7a–c. In Table 7a, the median
values, the MLE of α0 , α1 , β1 , γ1 and ν are given. We find that α0 is very close to
0 in all 15 models and the α1 in Models Rn and Rt (associated with the RV, r vt,i−1 2 )
is generally larger than the α1 in Models Gn and Gt (associated with the squared
2
return, rt,i−1 ). The difference in α1 among the models in Group 2 and Group 3 is more
pronounced when we incorporate seasonal indexes. These findings have an important
indication: RV may have greater explanatory power than return in the prediction of
volatility. The estimates of β1 in Group 2 are smaller than those in Group 1, indicating
that part of the effect of σt,i−1
2 on σt,i
2 can be attributed to the seasonal variation, which

is captured by S(i) in Group 2 models. A comparison between Group 2 and Group


3 results shows that the other parameter estimates do not change much even though
we allow time-varying degrees of freedom in Group 3. Consistent with past findings
using daily return, we also find that α1 + β1 is very close to 1 in all models. However,
the degrees of freedom are much smaller than those obtained with daily return series
and they are very close to the lower bound of 4 in Group 2. The small degrees of
freedom correspond to the large kurtosis of the intraday return and realized volatility,
suggesting the presence of dramatic price movements during the day.
Table 7b shows the median value of seasonal indexes in eight intraday intervals.
Consistent with the most volatile interval at 10:00–10:30 (i = 1) indicated by the
return and the realized volatility distribution in Sect. 2, seasonal index peaks after the
market opening, followed by a second peak at 14:30–15:00 (i = 6) after the lunch
break. The overall intraday volatility pattern follows a significant W-shape, as dis-
played in Fig. 3. The second peak of seasonal index at interval 6 can be attributed
to the after-lunch 30-min return, which accounts for the information released and the
market movements in Shenzhen and Shanghai during lunchtime in Hong Kong. On the
contrary, the median seasonal indexes have the lowest values at 10:30–11:00 (i = 2)
and 15:00–15:30 (i = 7). The change in variation from 15:00–15:30 to 15:30–16:00,
which is expressed as S(8)–S(7), is substantial. The market seems to enter from a
very volatile 30-min period right after lunch, to a ‘quiet’ 30-min period at 15:00–
15:30, and then to a volatile 30-min period before closing. The change from S(7) to
S(8) may be due the typical trading pattern of investors, especially day traders, who
tend to execute a large number of trades in the last 30 min to rebalance their portfo-
lios, to take profit or cut losses and to incorporate all the daily information in their

123
Forecasting Intraday Volatility and Value-at-Risk 97

Table 7 Results of parameter estimation for modified GARCH(1,1) models

Model α0 α1 β1 γ1 ν

a. Median values of basic GARCH parameters for 30 stocks (for Groups 1, 2 and 3)
Group 1
1Gn 0.006 0.052 0.946 / /
1Rn 0.005 0.054 0.935 / /
1GRn 0.007 0.011 0.927 0.048 /
1Gt 0.009 0.050 0.946 / 4.000
1Rt 0.003 0.071 0.914 / 4.000
1GRt 0.008 0.020 0.909 0.049 4.000
Group 2
2Gn 0.010 0.086 0.903 / /
2Rn 0.005 0.113 0.865 / /
2GRn 0.008 0.030 0.854 0.101 /
2Gt 0.015 0.110 0.884 / 4.000
2Rt 0.006 0.139 0.832 / 4.000
2GRt 0.009 0.044 0.810 0.113 4.492
Group 3
3Gt 0.013 0.110 0.886 / /
3Rt 0.008 0.156 0.825 / /
3GRt 0.010 0.039 0.818 0.127 /

i 1 2 3 4 5 6 7 8

b. Median values of the seasonal indexes S(i) for 30 stocks (for Groups 2 and 3)
Group 2
2Gn 2.732 0.278 0.751 0.912 0.961 2.563 0.660 1.196
2Rn 2.997 0.280 0.737 0.897 0.959 2.546 0.659 1.197
2GRn 2.970 0.276 0.741 0.905 0.954 2.501 0.652 1.184
2Gt 2.942 0.265 0.749 0.880 1.044 2.491 0.614 1.350
2Rt 2.978 0.267 0.705 0.822 1.028 2.453 0.626 1.309
2GRt 3.001 0.267 0.714 0.845 1.029 2.499 0.635 1.307
Group 3
3Gt 2.584 0.287 0.813 0.887 0.923 2.339 0.673 1.185
3Rt 2.705 0.298 0.759 0.923 0.993 2.172 0.733 1.219
3GRt 2.701 0.294 0.776 0.916 0.952 2.162 0.680 1.173
c. Median values of the time-varying degrees of freedom, Vi (for Group 3)
3Gt 8.429 4.331 3.752 3.518 4.216 4.503 4.039 4.837
3Rt 9.212 4.633 4.038 3.330 3.876 5.012 3.864 5.867
3GRt 9.693 4.544 4.059 3.344 3.811 4.784 4.159 6.360
i = 1, 2, 3, 4, 5, 6, 7, 8 correspond to the intraday intervals of 10:00–10:30, 10:30–11:00, 11:00–11:30,
11:30–12:00, 12:00–12:30, 14:30–15:00, 15:00–15:30, 15:30–16:00, respectively

strategies, etc. In summary, the seasonal indexes can be arranged in the following
order: S(1) > S(6) > S(8) > S(5) > S(4) > S(3) > S(7) > S(2). This order holds
in all the Group 2 and Group 3 models that we have investigated.

123
98 M. K. P. So, R. Xu

Fig. 3 W-shaped seasonal pattern for intraday conditional variance

In Table 7c, we report the median values of νi for the 30 stocks in Group 3 models.
From the table, most of the ν4 and ν5 in Models 3Gt , 3Rt and 3GRt are smaller than 4,
which explain that the kurtosis of the errors, εt,i , does not exist. Intervals 3–5, despite
having relatively small S(3), S(4) and S(5), are more fat-tailed than the other inter-
vals. This indicates extreme tail behavior of 30-min returns even after standardization
by the conditional variance. With the strong intraday seasonal effect reflected by S(1)
and S(8) in Table 7b, interval 1 has the highest degrees of freedom, or is the least
fat-tailed, followed by interval 8. This suggests that much of the tail risk in intervals 1
and 8 has been accounted for by the intraday volatility. In short, the seasonal dynamic
in the volatility and in the tail behavior give us a lot of insights into understanding the
risk of short-term portfolios.

3.3 Intrinsic Tail Risk Index

To further understand the impact of S(i) and νi on the evolution of risk in the eight
30-min intervals, we introduce the concept of the tail risk index below using Value at
Risk. Intraday VaR (IVaR) with confidence level 1 − α is defined by

Pr(rt,i < −IVaRt,i (α)|t,i−1 ) = α,

where rt,i is the realized 30-min return defined in Sect. 2; common confidence levels
used in the literature (e.g. Giot 2005) are 1−α = 95, 97.5, 99 and 99.5 %. We consider
all four confidence levels in our empirical study. Various approaches can be used to
forecast VaR, including parametric, nonparametric and semi-parametric. We use the
parametric calculation formula defined as follows:

IVaRt,i (α) = −σt,i Tν−1


i
(α), when εt,i ∼ T (0, 1, νi ), (4)

123
Forecasting Intraday Volatility and Value-at-Risk 99

where Tν−1 i
(α) is the αth percentile of T (0, 1, νi ). The formula in (4) allows the
degrees of freedom to vary and so is applicable to Group 3 models. In particular, if
either εt,i ∼ N (0, 1) or νi is static, we can set νi to ∞ and ν, respectively. From
2 = S(i) × τ 2 , we interpret the first factor, τ 2 , as
the decomposition in (3), i.e. σt,i t,i t,i
the explained return variation attributed to previous return movement and the second
factor, S(i), as an intrinsic scale factor independent of the market movement. The
IVaR in (4) can be rewritten as

IVaRt,i (α) = − S(i)τt,i Tν−1
i
(α),

from which the αth intrinsic tail risk index (TRI) is defined as

TRI(α) = − S(i) Tν−1
i
(α). (5)

The TRI represents the IVaR per unit of ‘explained price variation’, i.e. τt,i = 1. Since
the index does not depend on previous price information, it measures how bad the
return can be regardless of the market movement under extreme market stress. The
intraday tail risk, which is quantified by IVaR, is determined by the product of TRI in
(5) and τt,i , reflecting the latest market condition. Market participants can learn par-
tially from the TRI when the best time to trade is even without the most updated market
information. Table 8 presents the TRI(α) for Group 2 and 3 models at α = 0.5, 1, 2.5
and 5 %. The TRIs are obtained using the median S(i) and νi in Table 7. For the four
α we investigate, the ratio of the highest TRI recorded in interval 1 to the lowest TRI
recorded in interval 2 is about three. That means that the maximum loss in interval 1
as given by the IVaR can be three times as much as that in interval 2, provided that the
explained price variations τt,i in the two intervals are similar. For α = 1 %, the order
of TRI with respect to the intervals 1–8 is 1 > 6 > 8 > 5 > 4 > 3 > 7 > 2. The
order is stable across different α, meaning that the intrinsic tail risk is robust to the
confidence level we fix. From the order, the intrinsic tail risk increases from interval
2 (10:30–11:00) to interval 1 (10:00–10:30). The order of TRI exhibits a W-shaped
structure throughout the typical trading day. The W-shaped pattern for the model 3GRt
at four levels of α is representative of the three models in group 3 and displayed in
Fig. 4. In summary, our intrinsic tail risk index, which integrates the seasonality of
volatility and the tail behavior of returns, can give risk analysts a prior idea of the
intraday risk in specific trading intervals.

4 Intraday Value at Risk and Volatility Forecast Evaluation

4.1 Predicting Power of Absolute Returns

After estimating the parameters, we conduct out-of-sample forecasting of one-period-


ahead volatility using the validation data set from November 1st 2008 to May 31st
2009. Given that volatility is a latent variable, past literature use squared return, rt,i 2 , to

proxy for actual volatility r vt,i . Yet, as shown in Lopez (2001), while rt,i is an unbiased
2

estimator of σt,i
2 under (1) because of E[r 2 |
t,i t,i−1 ] = σt,i , it is very imprecise due to
2

123
100 M. K. P. So, R. Xu

Table 8 Intrinsic tail risk index of intervals 1–8 and α = 0.5, 1, 2.5 and 5 % obtained using the median
S(i) and in Table 7

i 1 2 3 4 5 6 7 8

α = 0.5 %
Group 2
2Gn 4.257 1.359 2.232 2.460 2.525 4.123 2.092 2.818
2Rn 4.459 1.364 2.211 2.441 2.522 4.110 2.091 2.819
2GRn 4.439 1.352 2.217 2.451 2.515 4.074 2.080 2.803
2Gt 5.584 1.656 2.788 3.054 3.272 5.102 2.549 3.772
2Rt 5.617 1.649 2.682 2.935 3.191 4.997 2.488 3.564
2GRt 5.594 1.631 2.682 2.911 3.166 5.016 2.495 3.573
Group 3
3Gt 4.572 1.685 2.984 3.058 3.043 4.717 2.735 3.381
3Rt 4.604 1.739 2.764 3.140 3.067 4.663 2.719 3.133
3GRt 4.599 1.693 2.764 3.090 3.059 4.646 2.606 3.074
α = 1%
Group 2
2Gn 3.845 1.228 2.016 2.222 2.280 3.724 1.890 2.545
2Rn 4.027 1.232 1.997 2.204 2.278 3.712 1.889 2.546
2GRn 4.009 1.222 2.002 2.213 2.272 3.679 1.879 2.531
2Gt 4.544 1.365 2.293 2.485 2.695 4.182 2.074 3.078
2Rt 4.572 1.358 2.212 2.388 2.642 4.108 2.064 2.976
2GRt 4.589 1.354 2.231 2.420 2.668 4.133 2.077 2.992
Group 3
3Gt 4.022 1.384 2.356 2.482 2.452 3.935 2.174 2.802
3Rt 4.093 1.433 2.231 2.514 2.548 3.859 2.248 2.725
3GRt 4.030 1.403 2.249 2.506 2.553 3.856 2.158 2.679
α = 2.5 %
Group 2
2Gn 3.239 1.035 1.699 1.872 1.921 3.138 1.593 2.144
2Rn 3.393 1.038 1.682 1.857 1.919 3.128 1.591 2.145
2GRn 3.378 1.029 1.687 1.865 1.914 3.100 1.582 2.133
2Gt 3.367 1.012 1.699 1.841 2.007 3.099 1.550 2.281
2Rt 3.415 1.021 1.649 1.798 2.009 3.115 1.557 2.252
2GRt 3.408 1.019 1.661 1.819 2.008 3.121 1.589 2.270
Group 3
3Gt 3.182 1.055 1.678 1.822 1.885 2.995 1.613 2.154
3Rt 3.281 1.078 1.681 1.840 1.935 2.931 1.632 2.162
3GRt 3.230 1.065 1.677 1.853 1.887 2.919 1.623 2.146
α = 5%
Group 2
2Gn 2.719 0.868 1.426 1.571 1.612 2.634 1.336 1.799
2Rn 2.847 0.871 1.412 1.558 1.611 2.625 1.335 1.800

123
Forecasting Intraday Volatility and Value-at-Risk 101

Table 8 continued

i 1 2 3 4 5 6 7 8

2GRn 2.834 0.864 1.416 1.565 1.607 2.602 1.328 1.790


2Gt 2.586 0.777 1.304 1.414 1.541 2.380 1.201 1.752
2Rt 2.671 0.799 1.266 1.399 1.556 2.424 1.226 1.749
2GRt 2.666 0.802 1.281 1.432 1.547 2.427 1.237 1.770
Group 3
3Gt 2.598 0.804 1.306 1.402 1.431 2.333 1.227 1.708
3Rt 2.642 0.840 1.285 1.388 1.517 2.306 1.232 1.692
3GRt 2.650 0.836 1.291 1.379 1.462 2.302 1.217 1.704

Fig. 4 W-shaped structure of intraday tail risk index for 2GRt

its asymmetric distribution. Christodoulakis and Satchell (1998) have also shown that
the mis-estimation of forecast performance is likely to be worsened by non-normality
which is known to be present in financial data. Hence, the use of rt,i 2 as a volatility

proxy may undermine the inference regarding forecast accuracy. Therefore, we pro-
pose two more measures to evaluate volatility forecast: |rt,i | and |rt,i |1.5 . According to
Forsberg and Ghysels (2007), an estimation of the variance function that is based on
absolute returns is more robust than if it is based on squared returns against asymmetry
or non-normality. The use of |rt,i |1.5 , however, is novel in our study. The motivation is
to draw a comparison between |rt,i | and rt,i 2 and make the evaluation more robust. We

also forecast one-period-ahead IVaR with the predicted volatility and conduct uncon-
ditional coverage test to evaluate IVaR forecasts. The calculation for the prediction of
powered return is specified below. For GARCH-normal models with rt,i = σt,i × εt,i
and εt,i ∼ N (0, 1), it can be shown that for c > 0, E[|rt,i |c | t,i−1 ] = σt,i c E[|Z |c ],

where Z is a standard normal random variable. To predict |rt,i |1.5 , it is natural to use
c E[|Z |c ], where σ c E[|Z |c ] can either have an explicit formula or can be computed
σt,i t,i

123
102 M. K. P. So, R. Xu

Fig. 5 Boxplot of 30-min predicted Conditional Variance (HSBC)

via numerical integration. For the popular choices of c = 1, 1.5 and 2, we determine
the following predictors for |rt,i |, |rt,i |1.5 and |rt,i |2 as

2
E[|rt,i | | t,i−1 ] = σt,i ,
π
E[|rt,i |1.5 | t,i−1 ] = σt,i
1.5
(0.86004), and
2
E[rt,i | t,i−1 ] = σt,i
2
.

For GARCH-t models with rt,i = σt,i × εt,i and εt,i ∼ T (0, 1, ν), we can derive
 ν−c 

E[|rt,i | | t,i−1 ] =
c
σt,i
c
(ν − 2) c/2
E[|Z | ]
c  ν2  2−c/2 . (6)
2

The proof of (6) can be found in the “Appendix”. In particular, we have


  
ν − 2 ν−1
E[|rt,i | | t,i−1 ] = σt,i   ,
2
π ν2

and


  ν−1.5
2
E[|rt,i |1.5 | t,i−1 ] = σt,i
1.5
(ν − 2)0.75 E |Z |1.5 ν  2−0.75 ,
2

where E[|Z |1.5 ] = 0.86004 and E[rt,i


2 |
t,i−1 ] = σt,i is the same as in the GARCH-
2

normal model.

123
Forecasting Intraday Volatility and Value-at-Risk 103

4.2 Forecasting Performance of Powered Intraday Returns

We show the boxplots of the eight 30-min interval conditional variances, σt,i 2 ,i =

1, 2, . . ., 8, for HSBC stock returns in the out-sample period, November 1st 2008 to
May 31st 2009 in Fig. 5. From the box plots, a clear seasonal pattern is revealed. As
reflected in S(1), the median conditional variance at 10:00–10:30 (i = 1) is the larg-
est among the eight intervals, followed by that at 14:30–15:00 after the lunch break.
Furthermore, the interquartile range of the conditional variance is also the largest at
10:00–10:30, indicating the volatility of volatility is at the highest right after the mar-
ket opening. Comparatively, the last interval, 15:30–16:00, has smaller median σt,i 2

than the after-lunch interval, 14:30–15:00, and the latter interval contains more wide-
spread extreme values making the conditional variance distribution highly skewed
to the right. On the other hand, the four intervals before lunch have smaller median
and spread, and the four distributions look quite similar. The conditional variance has
relatively low values and is relatively more stable at 11:00–12:30. Using the forecast
formula in Eq. (6), we calculate the predicted |rt,i |, |rt,i |1.5 , rt,i
2 and their mean abso-

lute error (MAE) to evaluate the 30-min volatility forecasting performance. The mean
absolute error of the predicted |rt,i |1.5 of the 30 stocks using the 15 models is pre-
sented in Table 9.3 Within each model, we observe large variations in MAE because
of the different volatility levels in different stocks. In many of the stocks we investi-
gate, the MAE from different models is quite stable. We highlight the best-performing
model(s) for each stock with bold formatting to identify consistently outperforming
models. It turns out that a majority of the best-performing models contain realized
volatility as an exogenous variable in the conditional variance equations (e.g. mod-
els 1Rt , 1GRt , 2Rt , 2GRt , 3Rt and 3GRt ). Most of the best models belong to class 2,
meaning that S(i) plays a more important role than time-varying degrees of freedom,
νi , in predicting |rt,i |1.5 .
In order to summarize the forecasting performance of |rt,i |, |rt,i |1.5 , and rt,i
2 in terms

of MAE, we follow So and Yu (2006) to produce a table of performance rankings.


For each stock, we rank the 15 models from 1 (smallest MAE) to 15 (largest MAE).
The performance rankings are then obtained by averaging the 30 ranks of each model.
By this construction, the lower the performance ranking, the smaller the MAE, which
indicates a better forecasting model. Table 10 presents the ranking results. Values less
than five are highlighted in grey with the smallest ranking in bold. For the forecasting
of |rt,i | and |rt,i |1.5 , t-error models are preferred, whereas for predicting rt,i
2 , normal-

error models do better. In terms of the prediction accuracy of |rt,i | and |rt,i |1.5 , model
2GRt performs best; for |rt,i |2 , model 2GRn gives the best prediction. Models with
realized volatility in building σt,i 2 usually give smaller MAE. This finding shows that

RV and seasonal index possess great explanatory power and can provide more accurate
volatility forecasts.

3 Tables for |r | and r 2 are also available upon request.


t,i t,i

123
104

Table 9 Mean absolute error of the predicted |rt,i |1.5 of the 30 stocks using the 15 models

123
1Gn 1Rn 1GRn 1Gt 1Rt 1GRt 2Gn 2Rn 2GRn 2Gt 2Rt 2GRt 3Gt 3Rt 3GRt

CK 0.78 0.76 0.76 0.76 0.76 0.76 0.76 0.73 0.73 0.75 0.76 0.73 0.76 0.72 0.73
CLP 0.39 0.44 0.40 0.35 0.41 0.37 0.34 0.37 0.35 0.32 0.35 0.34 0.33 0.35 0.34
Wharf 1.32 1.22 1.25 1.15 1.11 1.10 1.21 1.11 1.12 1.17 1.07 1.10 1.17 1.09 1.11
HSBC 0.52 0.48 0.49 0.47 0.46 0.47 0.48 0.43 0.44 0.46 0.43 0.44 0.46 0.43 0.43
HK Electric 0.48 0.51 0.49 0.43 0.45 0.44 0.44 0.46 0.44 0.42 0.44 0.42 0.43 0.49 0.43
HS Bank 0.74 0.81 0.81 0.69 0.73 0.73 0.67 0.72 0.71 0.65 0.69 0.69 0.67 0.70 0.72
Henderson 1.15 1.05 1.05 1.05 0.99 0.99 1.11 0.97 1.00 1.09 0.98 1.00 1.09 0.97 1.00
Hutchison 0.76 0.74 0.74 0.73 0.73 0.73 0.73 0.71 0.71 0.72 0.73 0.71 0.74 0.71 0.71
SHK 0.91 0.89 0.89 0.84 0.84 0.84 0.82 0.80 0.80 0.82 0.80 0.81 0.81 0.82 0.82
New World 1.19 1.25 1.23 1.10 1.16 1.13 1.09 1.10 1.09 1.04 1.12 1.07 1.06 1.10 1.09
Swire A 0.85 0.90 0.87 0.82 0.85 0.83 0.84 0.89 0.87 0.83 0.89 0.88 0.81 0.85 0.85
MTR 0.64 0.65 0.64 0.58 0.60 0.59 0.57 0.57 0.57 0.57 0.57 0.57 0.56 0.57 0.57
Sino Land 1.21 1.21 1.21 1.15 1.16 1.16 1.16 1.12 1.12 1.14 1.12 1.11 1.13 1.10 1.10
Hang Lung 1.37 1.45 1.30 1.19 1.18 1.24 1.18 1.15 1.16 1.13 1.11 1.11 1.11 1.15 1.11
CITIC Pacific 1.59 1.56 1.69 1.39 1.45 1.39 1.38 1.59 1.39 1.28 1.33 1.29 1.28 1.34 1.34
Cathay Pacific 1.11 1.12 1.12 1.06 1.00 0.99 0.93 0.98 0.98 0.92 0.95 0.94 0.90 0.96 0.97
Sinopec 0.89 0.91 0.91 0.85 0.87 0.86 0.87 0.90 0.90 0.87 0.88 0.88 0.86 0.88 0.88
HKEx 0.89 0.91 0.90 0.86 0.87 0.86 0.84 0.86 0.85 0.81 0.82 0.82 0.84 0.81 0.87
Li & Fung 1.30 1.34 1.34 1.08 1.03 1.10 1.15 1.12 1.12 1.06 1.07 1.07 1.10 1.11 1.15
China Unicom 0.92 0.93 0.93 0.84 0.85 0.85 0.88 0.88 0.88 0.86 0.86 0.85 0.83 0.87 0.85
PetroChina 0.76 0.77 0.77 0.74 0.74 0.74 0.74 0.73 0.74 0.74 0.72 0.72 0.72 0.72 0.72
CNOOC 0.89 0.85 0.85 0.87 0.85 0.84 0.85 0.80 0.80 0.84 0.79 0.79 0.83 0.79 0.79
M. K. P. So, R. Xu
Table 9 continued

1Gn 1Rn 1GRn 1Gt 1Rt 1GRt 2Gn 2Rn 2GRn 2Gt 2Rt 2GRt 3Gt 3Rt 3GRt

CCB 0.82 0.89 0.89 0.77 0.88 0.86 0.76 0.77 0.77 0.76 0.77 0.77 0.76 0.79 0.79
China Mobile 0.55 0.54 0.54 0.51 0.55 0.53 0.51 0.52 0.52 0.50 0.51 0.51 0.49 0.50 0.50
ICBC 0.59 0.59 0.59 0.58 0.59 0.58 0.55 0.54 0.54 0.55 0.54 0.54 0.56 0.54 0.54
FIH 1.99 2.17 2.08 1.83 1.78 1.90 1.92 2.07 1.98 1.88 2.05 1.96 1.90 2.01 2.01
Ping An 1.04 1.11 1.07 0.97 1.06 1.00 0.97 1.04 1.01 0.95 0.99 0.96 0.93 0.98 0.95
CHALCO 1.54 1.39 1.40 1.46 1.38 1.40 1.38 1.31 1.32 1.41 1.32 1.33 1.36 1.30 1.32
China Life 0.57 0.59 0.58 0.56 0.58 0.58 0.55 0.57 0.56 0.56 0.57 0.57 0.55 0.57 0.57
BOC 0.58 0.58 0.58 0.56 0.57 0.56 0.54 0.55 0.55 0.54 0.54 0.54 0.53 0.56 0.54
Bold values refer to the best performing model in terms of MAE
Forecasting Intraday Volatility and Value-at-Risk

123
105
106

123
Table 10 Volatility and IVaR forecasting performance rankings for the 15 models

1Gn 1Rn 1GRn 1Gt 1Rt 1GRt 2Gn 2Rn 2GRn 2Gt 2Rt 2GRt 3Gt 3Rt 3GRt
|rt,I | 13.67 10.13 13.97 9.47 10.50 10.20 9.23 9.20 8.33 4.00 4.83 3.80 3.83 4.70 4.13
Vol.
Rank |rt,I |1.5 13.03 13.87 13.57 7.87 9.63 8.70 7.57 7.47 6.90 5.10 5.73 4.57 4.70 5.90 5.40
rt,i2 8.77 9.67 9.27 7.10 9.23 8.03 4.87 4.63 3.73 9.57 8.97 7.20 9.80 9.87 9.30
0.5% 12.83 10.80 11.87 5.37 6.53 5.37 9.83 8.07 8.30 6.33 5.67 5.10 4.70 5.67 4.30
IVaR 1% 10.83 9.47 10.33 6.00 8.27 6.13 9.70 8.50 8.33 6.17 6.03 5.53 4.57 6.30 5.63
Rank 2.5% 7.33 8.20 7.77 8.63 9.67 7.30 7.07 8.67 7.37 5.17 7.00 5.77 5.90 8.13 7.00
5% 6.53 8.70 7.23 8.47 8.73 8.17 6.20 7.00 6.47 7.90 8.10 7.73 8.60 8.23 7.30

Bold values refer to the smallest ranking model


M. K. P. So, R. Xu
Forecasting Intraday Volatility and Value-at-Risk 107

4.3 Intraday VaR Predictive Performance

An unconditional coverage test of the predicted IVaR (Christoffersen 1998) is


also conducted to evaluate the IVaR forecasting performance. Based on the val-
ues of the likelihood ratio test statistics of the coverage test, we rank the models
for each stock and produce the performance rankings in Table 10 by averaging
the rankings among stocks. We highlight all rankings below seven and the best-
performing models as in Sect. 4.2. For 0.5 % IVaR, both the seasonal factor S(i)
and the time-varying degrees of freedom are important. Out of the best nine indi-
cated (1Gt , 1Rt . 1GRt , 2Gt , 2Rt . 2GRt , 3Gt , 3Rt , 3GRt ) for α = 0.5 %, models with
realized volatility are superior. Models 3GRt , 3Gt and 2Gt perform the best at
α = 0.5, 1 and 2.5 %, respectively. The results show that the time-varying degrees
of freedom in Models 3Gt , 3Rt and 3GRt help improve the prediction of 0.5 and 1 %
IVaR. For 5 % IVaR, model 2Gn outperforms the rest. To examine the statistical prop-
erties of empirical coverage, i.e. the proportion of 30-min returns smaller than their
IVaR, we display the distributions of the coverage using boxplots in Fig. 6a–d. It is
clear from Fig. 6a that normal-error models substantially underestimate IVaR (empir-
ical coverage > α), whereas moderate overestimation is recorded in t-error models
when α = 0.5 %. For the six models, 2Gt , 2Rt . 2GRt , 3Gt , 3Rt , 3GRt , the empiri-
cal coverage distributions are very much in coherence. In Fig. 6b, we can see that
most of the coverage of Models 2Gt and 3Gt (best performers) are within 1 ± 0.5%.
As in α = 0.5 %, the t-error models produce less biased and more stable cover-
age than normal-error models. In Fig. 6c, all models suffer from underestimation
of IVaR, yet models with t-error still outperform models with normal-error. Model
2Gt provides the most robust prediction. In Fig. 6d, the six normal-error models
have median coverage smaller than the nominal value of 5 %. The best perform-
ing model 2Gn gives symmetric coverage round α = 5 %. In short, the aggregate
model rankings are different for volatility forecasting and IVaR forecasting. In IVaR
prediction, there is no evidence that RV can improve the forecasting performance.
Models with time-varying degrees of freedom, νi , work particularly well in 0.5 and
1 % IVaR.
We can conclude from the above findings that t-error distribution works well for
0.5, 1 and 2.5 % VaR, which is different from the results in Giot (2005). In their study,
when α is at the 1, 2.5 and 5 % levels, the performance of models based on normal
distribution is quite satisfactory; the GARCH-t is rejected twice at the 5 % level. The
difference at α = 1 and 2.5 % may be attributed to two reasons. First, in Giot (2005),
the data set is limited to three stocks, BOEING, EXXON and IBM of the NYSE, in
a shorter period of 5 months. Their results, therefore, may not be extended to other
stocks in the NYSE. In our study, the data cover 30 HSI constituent stocks in a period
of over one year. Hence our findings are more representative of individual stocks listed
in Hong Kong. Second, stocks listed in these two markets may have different charac-
teristics and return distributions. Compared with BOEING, EXXON and IBM, most
of the 30 stocks analyzed in our study have a larger kurtosis for 30-min return, due
to relatively lower liquidity. Therefore, it is reasonable for the t-error distribution to
work better for all levels in our study.

123
108 M. K. P. So, R. Xu

(a)

1Gn 1Rn 1GRn 1Gt 1Rt 1GRt 2Gn 2Rn 2GRn 2Gt 2Rt 2GRt 3Gt 3Rt 3GRt

(b)

1Gn 1Rn 1GRn 1Gt 1Rt 1GRt 2Gn 2Rn 2GRn 2Gt 2Rt 2GRt 3Gt 3Rt 3GRt

Fig. 6 Boxplots for IVaR empirical coverage of the 15 models. a 0.5 % IVaR empirical coverage of the
15 models. b 1 % IVaR empirical coverage of the 15 models. c 2.5 % IVaR empirical coverage of the 15
models. d 5 % IVaR empirical coverage of the 15 models

5 Conclusion

In this paper, we first investigate the (properties of) intraday equity returns and real-
ized volatilities of 30 constituent stocks of the Hang Seng Index. We find that both
return and realized volatility demonstrate significant fat-tailed behavior. Consistent
with previous papers by Andersen and Bollerslev (1998), Giot (2000) and Giot (2005),
the empirical returns of the 30 stocks also feature a strong intraday seasonality in the
volatility: the price movement is most volatile after the market opens at 10:00–10:30
and after the lunch break at 14:30–15:00. This seasonal pattern is considered in our
volatility estimation model later. After standardizing return and taking logarithm of
realized volatility, the distributions become approximately normal in most cases.
Then we construct three groups of modified GARCH(1,1) models to predict one-
period-ahead volatility and intraday VaR. Although VaR models are usually applied

123
Forecasting Intraday Volatility and Value-at-Risk 109

(c)

1Gn 1Rn 1GRn 1Gt 1Rt 1GRt 2Gn 2Rn 2GRn 2Gt 2Rt 2GRt 3Gt 3Rt 3GRt

(d)

1Gn 1Rn 1GRn 1Gt 1Rt 1GRt 2Gn 2Rn 2GRn 2Gt 2Rt 2GRt 3Gt 3Rt 3GRt

Fig. 6 continued

to daily data to help manage the financial risks of banking and financial institutions,
we use the parametric VaR models to assess market risk on 30-min intraday returns.
Our time horizon is thus much shorter than what is usually considered in past VaR
literature, and is therefore more applicable to market participants involved in frequent
trading, such as market makers or day traders. In our modified GARCH(1,1) mod-
els, we incorporate RV and address the seasonality in both volatility and kurtosis by
introducing multiplicative seasonal indexes and time-varying degrees of freedom. The
models are applied to intraday data for the 30 stocks and they are ranked according to
their forecasting performance of intraday volatility and VaR. The overall performance
ranking shows that the Student t GARCH model with seasonal index and an additional
RV term performs the best in volatility forecasting. This indicates that RV can help
predict intraday volatility. Yet when it comes to IVaR prediction, RV does not seem
to help; Student t GARCH models with seasonal index and time-varying degrees of
freedom performs best for 0.5, 1 and 2.5 % IVaR; for 5 % VaR, the normal GARCH

123
110 M. K. P. So, R. Xu

model with seasonal index outperforms the others. Further investigation is worth to
understand why RV can help predict volatility but not IVaR. In summary, our results
demonstrate unequivocally that seasonal index, time-varying degrees of freedom and
t-error distribution can improve both volatility and IVaR prediction.

Appendix: Proof of Equation (6)



ν
Let X be a standard t random variable distributed as T (0, ν−2 , ν). We can represent
the distribution of X as a scaled mixture of normal, i.e.

X = s −1 Z ,

where νs 2 ∼ χν2 , a chi-square distribution with ν degrees of freedom, Z ∼ N (0, 1)


and s 2 and Z are independent. For any c < ν,
 
E[s −c ] = E ν c/2 (νs 2 )−c/2

1
= ν c/2 x −c/2  ν  ν/2 x ν/2−1 e−x/2 d x
2 2

1
= ν c/2  ν  ν/2 x −c/2 x ν/2−1 e−x/2 d x
2 2
 ν−c  (ν−c)/2
c/2 2 2

ν 2ν/2
 ν−c2 −c/2
2 2
= ν c/2   . (A.1)
ν2

The second equality is based on the distribution νs 2 ∼ χν2 and the second last
equality is obtained from the normalization constant of a χν−c
2 distribution. Therefore,

for GARCH-t models with rt,i = σt,i × εt,i and εt,i ∼ T (0, 1, ν), we have
   
E |rt,i |c | t,i−1 = σt,i
c
E |εt,i |c | t,i−1
  
 ν − 2 c
 
= σt,i
c
E  X  | t,i−1
 ν 
 c/2
ν−2
= σt,i
c
E[|s −1 Z |c | t,i−1 ]
ν

ν − 2 c/2 c  
= σt,i E[|Z |c ]E s −c | t,i−1
ν
 ν−c 
c  2  −c/2
= σt,i (ν − 2) E[|Z | ]
c c/2
2 .
ν2

The last equality above is obtained by substituting (A.1).

123
Forecasting Intraday Volatility and Value-at-Risk 111

References

Andersen, T. G., & Bollerslev, T. (1998). Answering the skeptics: Yes, standard volatility models do
provide accurate forecasts. International Economic Review, 39, 4.
Andersen, T. G., Bollerslev, T., Diebold, F., & Ebens, H. (2001a). The distribution of stock return
realized volatility. Journal of Financial Economics, 61, 43–76.
Andersen, T. G., Bollerslev, T., Diebold, F., & Labys, P. (2001b). The distribution of realized exchange
rate volatility. Journal of the American Statistical Association, 96, 42–55.
Barndorff-Nielsen, O. E., & Shephard, N. (2002). Econometric analysis of realised volatility and its use
in estimating stochastic volatility models. Journal of the Royal Statistical Society, Series B, 64,
253–280.
Beltratti, A., & Morana, C. (1999). Computing Value at Risk with high frequency data. Journal of
Empirical Finance, 6, 431–455.
Brownlees, C., & Gallo, G. (2010). Comparison of volatility measures: A risk management perspec-
tive. Journal of Financial Econometrics, 8, 29–56.
Christodoulakis, G. A., & Satchell, S. E. (1998). Hashing GARCH: A reassessment of volatility forecast
and performance. Forecasting Volatility in the Financial Markets, 168–192.
Christoffersen, P. F. (1998). Evaluating interval forecasts. International Economic Review, 39, 841–862.
Coroneo, L., & Veredas, D. (2006). Intradaily seasonality of returns distribution. A quantile regression
approach and intradaily VaR estimation. CORE Discussion Papers, 2006077.
Dionne, G., Duchesne, P., & Pacurar, M. (2009). Intraday Value at Risk (IVaR) using tick-by-tick data
with application to the Toronto stock exchange. Journal of Empirical Finance, 16, 777–792.
Fleming, J., Kirby, C., & Ostdiek, B. (2003). The economic value of volatility timing using realized
volatility. Journal of Financial Economics, 67, 473–509.
Forsberg, L., & Ghysels, E. (2007). Why do absolute returns predict volatility so well? Journal of
Financial Econometrics, 5, 31–67.
Fuertes, A. M., Izzeldin, M., & Kalotychou, E. (2009). On forecasting daily stock volatility. International
Journal of Forecasting, 25, 259–281.
Hansen, P. R., & Lunde, A. (2005). A forecast comparison of volatility models: Does anything beat a
GARCH(1,1)?. Journal of Applied Econometrics, 20(7), 873–889.
Hansen, P. R., & Lunde, A. (2006). Realized variance and market microstructure noise. Journal of
Business and Economic Statistics, 24, 127–161.
Giot, P. (2000). Time transformations, intraday data and volatility models. Journal of Computational
Finance, 4, 31–62.
Giot, P. (2005). Market risk models for intraday data. The European Journal of Finance, 11, 309–324.
Giot, P., & Laurent, S. (2004). Modeling daily Value-at-Risk using realized volatility and arch type
models. Journal of Empirical Finance, 11, 379–398.
Koopman, S. J., Jungbacker, B., & Hol, E. (2005). Forecasting daily variability of the S&P 100 stock index
using historical, realised and implied volatility measurements. Journal of Empirical Finance, 12,
445–475.
Lopez, J. A. (2001). Evaluating the predictive accuracy of volatility models. Journal of Forecasting,
20, 87–109.
Martens, M. (2001). Forecasting daily exchange return volatility using intraday returns. Journal of
International Money and Finance, 20, 1–23.
Martens, M. (2002). Measuring and forecasting S&P 500 index-futures volatility using high-frequency
data. Journal of Futures Markets, 22, 497–518.
So, M. K. P., & Yu, P.L.H. (2006). Empirical analysis of GARCH models in VaR estimation. Journal
of International Financial Markets, Institutions and Money, 16, 180–197.
Takahashi, M., Omori, Y., & Watanabe, T. (2009). Estimating stochastic volatility models using daily returns
and realized volatility simultaneously. Computational Statistics and Data Analysis, 53, 2404–2426.
Taylor, S. J., & Xu, X. (1997). The incremental volatility information in one million foreign exchange
quotations. Journal of Empirical Finance, 4, 317–340.
Zhang, L., Mykland, P., & Aït-Sahalia, Y. (2005). A tale of two time series: Determining integrated volatility
with noisy high-frequency data. Journal of the American Statistical Association, 100, 1394–1411.
Zhou, B. (1996). High-frequency data and volatility in foreign exchange rates. Journal of Business and
Economic Statistics, 14, 45–52.

123

You might also like