Transportation Research Part C 43 (2014) 5064

Contents lists available at ScienceDirect

Transportation Research Part C

journal homepage: www.elsevier.com/locate/trc

Adaptive Kalman lter approach for stochastic short-term

trafc ow rate prediction and uncertainty quantication
Jianhua Guo a,, Wei Huang a,1, Billy M. Williams b,2
a
b

Intelligent Transportation System Research Center, Southeast University, Si Pai Lou #2, Nanjing 210096, PR China
Department of Civil, Construction, and Environmental Engineering, North Carolina State University, Raleigh, NC 27695, USA

a r t i c l e

i n f o

Article history:
Received 17 June 2013
Received in revised form 26 January 2014
Accepted 13 February 2014

Keywords:
Congestion
Intelligent transportation system
Short term trafc ow forecasting
SARIMA
GARCH
Adaptive Kalman lter

a b s t r a c t
Short term trafc ow forecasting has received sustained attention for its ability to provide
the anticipatory trafc condition required for proactive trafc control and management.
Recently, a stochastic seasonal autoregressive integrated moving average plus generalized
autoregressive conditional heteroscedasticity (SARIMA + GARCH) process has gained
increasing notice for its ability to jointly generate trafc ow level prediction and associated prediction interval. Considering the need for real time processing, Kalman lters have
been utilized to implement this SARIMA + GARCH structure. Since conventional Kalman
lters assume constant process variances, adaptive Kalman lters that can update the
process variances are investigated in this paper. Empirical comparisons using real world
trafc ow data aggregated at 15-min interval showed that the adaptive Kalman lter
approach can generate workable level forecasts and prediction intervals; in particular,
the adaptive Kalman lter approach demonstrates improved adaptability when trafc is
highly volatile. Sensitivity analyses show that the performance of the adaptive Kalman
lter stabilizes with the increase of its memory size. Remarks are provided on improving
the performance of short term trafc ow forecasting.
2014 Elsevier Ltd. All rights reserved.

1. Introduction
Congestion is causing serious issues for surface transportation systems around the world. Due to the increasing constraints on new road construction or expansion, trafc management and control systems under the umbrella of intelligent
transportation systems (ITS) have become increasingly vital for improving the efciency and safety of trafc operations. In
contrast to reactive management and control systems that respond to currently observed trafc conditions, the proactive
systems rely on accurate prediction of near-term trafc conditions. Considering the importance of trafc ow rate, dened
as the number of vehicles passing a specic road section over a predened time interval (TRB, 2000), short-term trafc ow
rate forecasting has been identied as one of the major challenges for developing proactive ITS applications.
Short term trafc ow rate forecasting includes trafc ow rate level prediction, i.e., point forecast, and uncertainty quantication associated with level prediction, i.e., prediction interval generation (Chateld, 1993). Intuitively, short-term trafc
ow rate forecasting should be informed by our understanding of the trafc ow rate dynamics that is of primary

Corresponding author. Tel.: +86 25 83793131.

1
2

E-mail addresses: jg2nh@yahoo.com (J. Guo), hhhwei@126.com (W. Huang), billy_williams@ncsu.edu (B.M. Williams).
Tel.: +86 25 83793131.
Tel.: +1 919 5157813.

http://dx.doi.org/10.1016/j.trc.2014.02.006
0968-090X/ 2014 Elsevier Ltd. All rights reserved.

J. Guo et al. / Transportation Research Part C 43 (2014) 5064

51

importance in forecasting algorithm development (Stephanedes et al., 1981). Recently, based on previous ndings in Williams (1999), Williams and Hoel (2003), Shekhar and Williams (2008), etc., a stochastic structure of seasonal autoregressive
integrated moving average plus generalized autoregressive conditional heteroscedasticity (SARIMA + GARCH) is utilized to
model the rst and second conditional moment of trafc ow rate series, and this structure has been shown to be able to
generate desirable trafc ow rate level forecasts and prediction intervals (Guo et al., 2008).
Due to the requirement of real time processing for many ITS-related transportation applications, the SARIMA + GARCH
structure needs to be handled in an online fashion. In this regard, Kalman lter, due to its recursive nature, is one of the most
widely adopted tools of achieving this purpose. In doing so, the time series models are converted into state space models,
including a state transition equation and an observation equation for the hidden state process and the observation process,
respectively. In the conventional Kalman lter, the process variances are important seeding parameters calibrated usually
using historical trafc ow rate series. However, as shown in Guo (2005) and Guo et al. (2012), trafc condition is heteroscedastic in nature, and adaptive Kalman lters with time-varying process variances adaptation are expected to be theoretically more appealing than the conventional Kalman lters. In this paper, the performance of the adaptive Kalman lter for
short term trafc ow rate forecasting is investigated. Following a review of related literatures, the methodology is presented, including the denition of the stochastic structure, state space representation, and adaptive Kalman recursion.
The paper then presents the empirical results from application of the adaptive Kalman lter to real world data. The paper
concludes with summaries and remarks.
2. Literature review
Over the decades, there has been a variety of approaches published on short term trafc ow forecasting. In this section,
we summarize the studies of level forecasting and uncertainty quantication, i.e., point forecast and prediction interval
generation.
2.1. Trafc condition level forecasting
Trafc ow level forecasting, i.e., point prediction, has been widely investigated and many studies have been published in
the literature. First, linear time series model has been widely applied, including exponential smoothing (Ross, 1982), autoregressive integrated moving average (ARIMA) model (Ahmed and Cook, 1979; Levin and Tsao, 1980; Nihan and Holersland,
1980; Hamed et al., 1995), SARIMA (Williams, 1999; Williams and Hoel, 2003; Guo, 2005; Guo et al., 2008), multivariate time
series models (Williams, 2001; Kamarianakis and Prastacos, 2003; Min and Wynter, 2011), spectral analysis (Nicholson and
Swann, 1974; Tchrakian et al., 2012). In these models, the trafc dynamics are inherently assumed to be linear. It is worthwhile to note that SARIMA has been shown to generate promising performances (Smith et al., 2003; Guo et al., 2008; Lippi
et al., 2013).
Second, non-parametric approaches have been applied, including neural network and its variations (Clark et al., 1992;
Smith and Demetsky, 1994; Park et al., 1998; Zhang, 2000; Dia, 2001; Chen and Grant-Muller, 2001; Yin et al., 2002; Jiang
and Adeli, 2005; Vlahogianni et al., 2005; Dunne and Ghosh, 2012), k-nearest neighbor approach (Davis and Nihan, 1991;
Smith and Demetsky, 1996, 1997), kernel smoothing (El Faouzi, 1996), and local linear regression (Sun et al., 2003). These
methods are in general automatic and do not make strong assumptions on the underlying model form. They have a notable
advantage of adaptive learning of the underlying trafc dynamics through a historical trafc condition database.
Hybrid methods have also been exploited to enhance the performances of single forecasting approaches, including combined Kohonen maps with ARIMA (KARIMA) model (Der Voort et al., 1996), ATHENA (Danech-Pajouh and Aron, 1991), Bayesian-neural network approach (Zheng et al., 2006), hybrid fuzzy rule-based approach (Dimitriou et al., 2008), hybrid EMDBPN (empirical mode decomposition-back propagation neural networks) approach (Wei and Chen, 2012), chaos-wavelet
analysis-support vector machine approach (Wang and Shi, 2013). Intuitively, the implementation characteristics of hybrid
methods are generally complex, thereby discouraging their wide-scale implementations.
In addition, ltering approaches have been widely applied, including recursive least square (RLS) (Kang et al., 1998; Yang
et al., 2004), Kalman lter (Gazis and Knapp, 1971; Okutani and Stephanedes, 1984; Stathopoulos and Karlaftis, 2003; Shekhar, 2004; Guo, 2005), generalized linear model (GLM) (Lan and Miaou, 1999; Lan, 2001), Bayesian dynamic linear model
(Fei et al., 2011), and least mean square (LMS) lters (Lu, 1990). It is worthwhile to note that the RLS, Kalman lter, and GLM
are closely related by assuming the random walk model for the forecasting algorithm state evolution (Yang et al., 2004; Lan
and Miaou, 1999; Lan, 2001). These methods are promising in imparting a self-adjusting ability into the forecasting process
and the Kalman lter approach is theoretically appealing for short term trafc condition forecasting. However, as pointed out
previously, the heteroscedastic nature of trafc condition series demands a process variances adaptation mechanism in the
Kalman lters.
2.2. Uncertainty quantication
Compared with trafc ow level forecasting, studies on trafc uncertainty quantication or prediction interval generation
are initially limited to several studies with discouraging results (Yang et al., 2004; Hugosson, 2005; Pattanamekar et al.,

52

J. Guo et al. / Transportation Research Part C 43 (2014) 5064

2003; Sun et al., 2004). Recently, models are borrowed from nancial analysis eld for uncertainty quantication with promising results, including primarily GARCH approach (Guo, 2005; Kamarianakis et al., 2005; Tsekeris and Stathopoulos, 2006;
Guo et al., 2008; Sohn and Kim, 2009; Karlaftis and Vlahogianni, 2009;Yang et al., 2010; Guo and Williams, 2010; Chen et al.,
2011; Zhang et al., 2013), and stochastic volatility approach (Tsekeris and Stathopoulos, 2010). Note that GARCH model can
be processed using Kalman lter (Guo, 2005; Guo et al., 2008; Guo and Williams, 2010); therefore, adaptive mechanism can
also be used for improving the performance. In addition, as a conventional prediction model, neural network has been enhanced to provide the ability of generating prediction intervals (van Hinsbergen et al., 2009; Khosravi et al., 2011; Mazloumi
et al., 2011).
2.3. Summary
In summary, many studies have been proposed for short term trafc ow level forecasting and prediction interval generation. Specically, Kalman lter approach developed based on time series models, in particular, the SARIMA + GARCH
structure, has been recognized as promising for short term trafc ow forecasting. However, conventional Kalman lter cannot update its process variances in real time. Therefore, this paper will investigate and test an adaptive Kalman lter approach using real world trafc ow data collected around the world.
3. Methodology
In this section, rst, the stochastic structure of SARIMA + GARCH is dened; then its state space representation is presented; nally the adaptive Kalman recursion is provided.
3.1. SARIMA + GARCH structure
As mentioned previously, the stochastic SARIMA + GARCH structure is gradually emerging as one of the promising candidates for modeling trafc ow series. In this structure, the SARIMA component captures the rst conditional moment,
i.e., the dynamics of trafc ow levels, and the GARCH component captures the second conditional moment, i.e., the dynamics of trafc ow variances. Formally, for a discrete trafc ow series xt, the SARIMA(p,d,q)(P,D,Q)S + GARCH(u,v) structure is
dened as Eqs. (1)(4).
D

/BUBS 1  BS 1  Bd xt hBHBS et

et

p
ht et

ht a0

1
2

v
X

u
X
bi hti

i1

i1

ai e2ti

et  IN0; 1
where
t: time index;
p the order of the short-term autoregressive polynomial;
q: order of the short-term moving average polynomial;
d: order of the short-term differencing;
P: order of the seasonal autoregressive polynomial;
Q: order of the seasonal moving average polynomial;
D: order of the seasonal differencing;
B: backshift operator such that Bxt = xt1;
(1  BS)D: seasonal differencing;
(1  B)d: short term differencing;
/(B) = 1  /1B  /2B2      /pBp: short-term autoregressive polynomial;
h(B) = 1  h1B  h2B2      hqBq: short-term moving average polynomial;
U(BS) = 1  U1(BS)  U2(BS)2      UP(BS)P: seasonal autoregressive polynomial;
H(BS) = 1  H1(BS)  H2(BS)2      HQ(BS)Q: seasonal moving average polynomial;
ht: conditional variance at t, i.e., et|Wt1  N(0, ht) with Wt1 as the information up to t  1;
u: autoregressive order of GARCH process with u P 0;
v: moving average order of GARCH process with v > 0;
a0: positive constant coefcient;
ai,i=1,. . .,v: non-negative coefcients of the lagged sample variance e2ti ;
bi,i=1,. . .,u: non-negative coefcients of the lagged conditional variance hti.

J. Guo et al. / Transportation Research Part C 43 (2014) 5064

53

In above representation, Eq. (1) denes the SARIMA structure, and Eqs. (2)(4) dene the GARCH structure. For the SARIMA structure, it is assumed that the roots of /(B), h(B), U(BS), and H(BS) are outside of unit circle to ensure the causality and
invertibility of the model, and also these polynomials have no common factors. Note that the causality property ensures the
process can be expressed in terms of past innovations and the invertibility property ensures the process can be expressed as
past observations, which is important for model estimation and prediction (Fuller, 1996; Box et al., 1994, 2008). et is the
residual series, conventionally assumed to be Gaussian white noise with mean zero and constant variance; however, for heteroscedastic trafc ow series as shown in (Guo et al., 2012), the addition of the GARCH component is needed. In Guo et al.
(2012), through comprehensive transformation analyses and heteroscedasticity tests with different testing powers, heteroscedasticity is shown to be intrinsic to trafc ow series, which validates the necessity of further investigations using approaches such as the GARCH model.
For the GARCH structure, the error series et is assumed to be normal for facilitating the construction of the prediction
intervals. An alternative representation establishes that GARCH process can be interpreted as autoregressive moving average
(ARMA) process in e2t as presented in Eq. (5).

e2t a0

n
u
X
X
ai bi e2ti 
bi gti gt
i1

i1

where gt e2t  ht that is serially uncorrelated with mean zero and n = max (u, v) (Bollerslev, 1986). Note that this ARMA
interpretation provides a handy treatment of converting the GARCH model to a state space representation.
Given the specication of the SARIMA + GARCH structure, the orders of SARIMA and GARCH can be determined, respectively. For the SARIMA structure, previous studies have shown that the SARIMA(1,0,1)(0,1,1)672 model can be used to describe trafc ow rate series aggregated at 15-min interval (Williams, 1999; Williams and Hoel, 2003). Based on the
Wold Decomposition Theorem and the assertion that a one-week lagged rst seasonal difference will yield a weakly stationary transformation, Williams (1999) and Williams and Hoel (2003) showed that trafc ow series can be modeled a SARIMA
process using real world trafc ow data collected from the United States and the United Kingdom. By investigating into this
SARIMA model, we can see that SARIMA(1,0,1)(0,1,1)672 model can be processed as a cascade of a seasonal operator of
IMA(0,1,1)672, i.e., the integrated moving average operator, and a short-term operator of ARMA(1,1), i.e., the autoregressive
moving average operator.
The seasonal operator is dened as Eq. (6), i.e.,

1  B672 xt 1  HB672 wt

where wt is the output of the seasonal operator. A further manipulation gives Eqs. (7) and (8) as

1  HB672  1  HB672
1  HB672

xt wt

and
2

xt 1  HB672 HB672 H2 B672   xt wt

indicating that the seasonal operator is exactly a seasonal exponential smoothing with smoothing parameter a = 1  H. In
fact, this seasonal operator will retrieve the seasonal pattern in trafc ow series.
The short-term operator is dened as Eq. (9), i.e.,

1  /Bwt 1 hBet

which models local variations after the seasonal variation is removed from the original trafc ow series through the
seasonal operator.
For the GARCH(u,v) process, although an exact best model can be identied by minimizing the sample information criterion, a parsimonious model is preferable for eld implementations. In this paper, the GARCH(1,1) model is selected for its
simplicity. For this model, Guo (2005) showed that the information criteria of the best model and the GARCH(1,1) model
are close to each other for multiple trafc ow series collected around the world, indicating the validity of choosing this
GARCH(1,1) model.
In summary, for the 15-min trafc ow data, the SARIMA(1,0,1)(0,1,1)672 + GARCH(1,1) structure can be utilized for short
term trafc ow level prediction and uncertainty quantication. Note that this structure can be processed in two steps, i.e.,
SARIMA processing and GARCH processing using outputs from SARIMA processing, without efciency lost (Engle, 1982; Bollerslev, 1986). This is essential for constructing the online short term trafc ow forecasting system.
3.2. State space representation
Recall that the SARIMA model can be interpreted as a cascade of a seasonal operator and a short-term operator; therefore,
the SARIMA (1,0,1)(0,1,1)672 + GARCH(1,1) structure can be processed as IMA(0,1,1)672 + ARMA (1,1) + GARCH(1,1), which
can be converted into a seasonal exponential smoothing operator plus two state space representations as follows.

54

J. Guo et al. / Transportation Research Part C 43 (2014) 5064

For IMA(1,0,1)672, recognizing its equivalence with seasonal exponential smoothing shown previously, the operator can
be processed as Eq. (10), i.e.,

^xt axt672 1  a^xt672

10

xt is the predicted value at time t. In this paper, we assume ^

xt xt
where a is the smoothing parameter with a = 1  H and ^
for t = 1, 2,. . ., 672 to start the equation.
For ARMA(1,1) + GARCH(1,1), Guo and Williams (2010) presented two state space representations for modeling and predicting trafc speed series. For the ARMA component, we reorganize the ARMA(1,1) model into Eq. (11), i.e.,

xt /xt1 het1 et

11

then by treating the ARMA model parameters as hidden state to be estimated and using the random walk model with a forgetting factor as its evolution dynamics, we dene the state transition equation as Eq. (12), i.e.,

wt Uwt1 at :

12

In addition, by treating the trafc ow rate series as the driving observation process, we dene the observation equation
as Eq. (13), i.e.,

Y t X Tt wt et

13

where
wt: state variable dened as / h T ;
1
U: state transition matrix dened as diagfk2 g, with k dened as a forgetting factor;
at: state noise series with state noise covariance matrix Q t Cov at aTt ;
Yt: current observation dened as xt;
Xt: time varying observation matrix dened as xt1 et1 T ;
et: observation noise series with observation noise covariance matrix Rt Cov et eTt .
Combined, Eqs. (12) and (13) constitute the state space representation for the ARMA(1,1) operator.
For the GARCH(1,1) component, according to Eq. (5) and through a reparameterization, we have Eq. (14) as

e2t a0 ae2t1 bgt1 gt

14

where b = b1 and a = a1 + b1. Then, we dene the observation equation as Eq. (15), i.e.,

e2t 1 e2t1

0 1
a0
B C
gt1 @ a A gt
b

15

and the state transition equation as Eq. (16), i.e.,

1
0 1
a0
n 1 o a0
B C
B C

@ a A diag k 2 @ a A zt
b
b
t
t1

16

where
gt: observation noise series with observation noise covariance matrix Cov gt gTt Rt ;
zt: state noise series with state noise covariance matrix Cov zt zTt Q t .
Combined, Eqs. (15) and (16) constitute the state space representation for GARCH(1,1) model, where the reparameterized
parameter vector a0 a b T is treated as the hidden state and the squared residual series from ARMA(1,1) model is treated
as the driving observation process.
Therefore, the SARIMA(1,0,1)(0,1,1)672 + GARCH(1,1) structure can be processed consecutively using the seasonal exponential smoothing operator, i.e., Eq. (10), and two state space representations, i.e., Eqs. (12) and (13) and Eqs. (15) and
(16). Since Eq. (10) is recursive in nature and the two state space models can be solved recursively, an online prediction system can then be constructed.
3.3. Adaptive Kalman recursion
Though the two state space models described above can be readily solved using the well-known Kalman recursions (Kalman, 1960), it is necessary to point out that the process variances required for seeding the Kalman recursion could be either
constant or time varying. For heteroscedastic trafc ow rate series, an adaptation mechanism for updating the process variances is preferred, termed as the so called adaptive Kalman lter. In this paper, the adaptation mechanism proposed in Myers
and Tapley (1976) is selected that uses a memory of observation errors and state estimation errors to ne-tune the process
variances. Using the state space representation in Eqs. (12) and (13) as an example, the steps of the adaptive Kalman recursion are described as Eqs. (17)(27).

J. Guo et al. / Transportation Research Part C 43 (2014) 5064

55

Step 1: State propagation and prior state estimation error covariance estimation

bw
^ tjt1 U
^ t1jt1
w

17

T
b UP
b
P
tjt1
t1jt1 U Q t

18

Step 2: Observation errors computation

^ tjt1
et Y t  X Tt w

19

Step 3: Update observation process covariance matrix Rt

^e

N
1X
etj1
N j1

Rt


N 
1X
N1
T
T
b
X tj1 P
etj1  ^eetj1  ^e 
X
tj1jtj tj1
N j1
N

20

21

Step 4: Kalman gain computation

Kt

b Xt
P
tjt1
b  X t Rt
XT P
t

22

tjt1

Step 5: Posterior state estimation and posterior state estimation error covariance estimation



^ tjt w
^ tjt1 K t Y t  X Tt w
^ tjt1
w

23



b I  K t XT P
b
P
tjt1
t
tjt

24

Step 6: State estimation errors computation

^ tjt  Uw
^ t1jt1
at w

25

Step 7: Update state process covariance matrix Qt

^
a

N
1X
atj1
N j1

Qk

N 
i



1X
N  1h
^ atj1  a
^ T
atj1  a
Utj1 Pb tjjtj UTtj1  Pb tj1jtj1
N j1
N

26

27

where
b  : prior state estimation error covariance;
P
tjt1
b : posterior state estimation error covariance;
P
tjt
Kt: Kalman gain at time t;
et: observation errors;
^et : average observation errors;
at: system state estimation errors;
^t : average system state estimation errors;
a
N: prescribed memory size of the adaptive Kalman lter.
3.4. Summary
In summary, the SARIMA(1,0,1)(0,1,1)672 + GARCH(1,1) structure can be used to model the 15-min trafc ow data, and
this structure can be converted into a seasonal operator and two state space representations. In this paper, an adaptive Kalman recursion mechanism will be applied onto the two state space representations so that two adaptive Kalman lters can
be developed to process sequentially the trafc ow series in real time.
4. Empirical study
In this section, we present the real world data used in this study, the experimental design for carrying out the investigation, the empirical results on level forecasting and uncertainty quantication, and a sensitivity analysis on the memory size
of the adaptive Kalman lters.

56

J. Guo et al. / Transportation Research Part C 43 (2014) 5064

4.1. Data
Real world trafc data collected from four highway systems are used in this paper, including the motorway system in the
United Kingdom, and the metropolitan freeway systems in Maryland, Minnesota, and Washington State of the United States.
Though originally archived at different time intervals, these data are aggregated into 15-min data according to Edie (1963),
i.e., the 15-min trafc ow data were computed as the sum of trafc ow data for all the original time intervals within this
15-min interval. In performing the aggregation, missing values were propagated upwards, and simple screening procedures,
i.e., threshold test and hang-on test, were applied to eliminate the obvious erroneous data. In addition, missing values were
imputed using the SARIMA(1,0,1)(0,1,1)672 model through applying the back forecasting technique iteratively. In total, 36
trafc ow series were selected and a summary of the data is provided in Table 1.
4.2. Experimental design
In order to investigate the performances of the proposed adaptive Kalman lter, four approaches are compared as below
in Table 2.
Note that in Table 2, for BATCH processing, the SARIMA model outperforms HoltWinters method, three heuristic models
(i.e., random walk, historical average, and derivation from historical average), neural network, and k-nearest neighbor
regression in terms of level forecasting in these studies (Williams, 1999; Smith et al., 2003; Williams and Hoel, 2003),
and the GARCH model is the primary approach that has been investigated in terms of uncertainty quantication. Therefore,
the comparison of the four selected approaches will provide a comprehensive picture on short term trafc ow rate
forecasting.
For each approach, three measures of level forecasting are used, i.e., mean absolute error (MAE), mean absolute percentage error (MAPE), and root mean square error (RMSE). Given xt as real trafc ow rate observations, ^
xt as the forecasted

Table 1
Data overview.
Region

Highway

Station

Number of lanes

Start

End

Sample size

UK
UK
UK
UK
UK
UK
UK
UK
UK
UK
UK
UK
MD
MD
MD
MD
MD
MD
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
WA
WA
WA
WA
WA
WA

M25
M25
M25
M25
M25
M25
M25
M25
M1
M1
M1
M6
I270
I95
I795
I795
I695
I695
I35W-NB
I35W-SB
I35E-NB
I35E-SB
169-NB
169-SB
I35W-NB
I35W-SB
I35E-NB
I35E-SB
169-NB
169-SB
I5
I5
I5
I5
I405
I405

4762a
4762b
4822a
4826a
4868a
4868b
4565a
4680b
2737a
2808b
4897a
6951a
2a
4b
7a
7b
9a
9b
60
578
882
890
442
737
60
578
882
890
442
737
ES-179D_MN_Stn
ES-179D_MS_Stn
ES-130D_MN_Stn
ES-130D_MS_Stn
ES-738D_MN_Stn
ES-738D_MS_Stn

4
4
4
4
4
4
4
4
3
3
3
3
3
4
2
2
4
4
4
3
3
3
2
2
4
3
3
3
2
2
4
3
4
4
3
3

9/1/1996
9/1/1996
9/1/1996
9/1/1996
9/1/1996
9/1/1996
1/1/2002
1/1/2002
2/13/2002
2/13/2002
2/13/2002
1/1/2002
1/1/2004
6/1/2004
1/1/2004
1/1/2004
1/1/2004
1/1/2004
1/1/2000
1/1/2000
1/1/2000
1/1/2000
1/1/2000
1/1/2000
1/1/2004
1/1/2004
1/1/2004
1/1/2004
1/1/2004
1/1/2004
1/1/2004
1/1/2004
4/1/2004
4/1/2004
7/1/2004
7/1/2004

11/30/1996
11/30/1996
11/30/1996
11/30/1996
11/30/1996
11/30/1996
12/31/2002
12/31/2002
12/31/2002
12/31/2002
12/31/2002
12/31/2002
5/5/2004
11/15/2004
5/5/2004
5/5/2004
5/5/2004
5/5/2004
12/31/2000
12/31/2000
12/31/2000
12/31/2000
12/31/2000
12/31/2000
12/31/2004
12/31/2004
12/31/2004
12/31/2004
12/31/2004
12/31/2004
6/29/2004
6/29/2004
9/30/2004
9/30/2004
12/29/2004
12/29/2004

8736
8736
8736
8736
8736
8736
35,040
35,040
30,912
30,912
30,912
35,040
12,096
16,128
12,096
12,096
12,096
12,091
35,136
35,136
35,136
35,136
35,136
35,136
35,136
35,136
35,136
35,136
35,136
35,136
17,298
17,298
17,516
17,516
17,406
17,406

Note: UK: the United Kingdoms; MD: Maryland; WA: Washington State; MN: Minnesota. See Guo et al. (2012) for more information.

57

J. Guo et al. / Transportation Research Part C 43 (2014) 5064

Table 2
Approaches under comparison.
Approach

Description

EXPRW

Uses seasonal exponential smoothing for capturing the historical pattern, and uses a random walk model to capture the local variations;
considering the established robustness of exponential smoothing, the seasonal exponential smoothing parameter a is selected as 0.15,
and so is for the following three approaches (Williams, 1999; Williams and Hoel, 2003; Shekhar, 2004; Guo, 2005). This method does not
support the construction of time varying prediction intervals
Uses SAS PROC ARIMA to process the SARIMA(1,0,1)(0,1,1)672 model according to the standard Box-Cox approach, and uses SAS PROC
AUTOREG to process the GARCH(1,1) model. Square root transformation is applied in the data series according to Guo (2005) and Guo
et al. (2012). Time varying prediction intervals are computed at 95% signicant level
Uses seasonal exponential smoothing and two standard Kalman lters for processing the SARIMA(1,0,1)(0,1,1)672 + GARCH(1,1) structure;
square root transformation is applied in the data series according to Guo (2005) and Guo et al. (2012). Time varying prediction intervals
are computed at 95% signicant level
Uses seasonal exponential smoothing and two adaptive Kalman lters for processing the SARIMA(1,0,1)(0,1,1)672 + GARCH(1,1) structure;
square root transformation is applied in the data series according to Guo (2005) and Guo et al. (2012). Time varying prediction intervals
are computed at 95% signicant level. The memory of each adaptive Kalman lter is selected as 672, i.e., one week

BATCH

KF

AKF

trafc ow rates, and n as the total number of trafc ow rate observations processed, the three measures are dened in Eqs.
(28)(30), respectively.

MAE

n
1X
jxt  ^xt j
n t1

MAPE

n
^
100 X
xt  xt

n t1 xt

RMSE

1
n

28

29

q
Xn
2
n t1 xt  ^xt

30

For each approach, except for EXPRW that cannot generate time varying prediction intervals, two measures of uncertainty
quantication performance are calculated, namely kickoff percentage and width to ow ratio. The kickoff percentage is calculated as the total number of trafc ow observations lying outside of prediction intervals divided by the total number of
trafc ow rate observations, and the width to ow ratio is calculated as the average of width to ow ratios for all the prediction intervals. Ideally, the kickoff percentage is expected to be close to 5% for 95% signicant level, and the width to ow
ratio is expected to be small.
In order to show the detailed performance of each approach, these measures will be investigated by groups according to
trafc ow levels and time of day as dened in Tables 3 and 4, respectively. Note that groups with fewer than 10 observations are ignored due to limited sample size. In addition, the mean performance measures will be computed for each measure across these trafc ow series.
4.3. Results on level forecasting
The results on level forecasting are presented in Table 5 for all the trafc ow observations. To compare these approaches
pair-wisely, a protected repeated t-test is used on the performance measures of MAE, MAPE, and RMSE over the 36 stations,
respectively, with same group code in Table 5 indicating no signicant statistical difference in the mean performances (SAS
Institute Inc., 1999). On observing Table 5, rst, for all the three measures, the statistically signicant mean performance difference between BATCH, AKF, and KF cannot be established through the repeated t-test, showing comparable mean performances of the three methods in level forecasting. In addition, the three methods outperformed EXPRW in that though the
seasonal exponential smoothing part of EXPRW has the theoretical foundation to capture the historical trafc ow rate pat-

Table 3
Groups by trafc level.
Group

Group description

TTL
G1
G2
G3
G4
G5

For all trafc observations

P0 and <500 veh/h/ln
P500 and <1000 veh/h/ln
P1000 and <1500 veh/h/ln
P1500 and <2000 veh/h/ln
P2000 veh/h/ln

58

J. Guo et al. / Transportation Research Part C 43 (2014) 5064

Table 4
Groups by time of day.
Group

Group description

T1
T2
T3
T4
T5
T6
T7
T8
T9
T10
T11
T12
T13
T14
T15

[00:0004:00)
[04:0006:00)
[06:0007:00)
[07:0008:00)
[08:0009:00)
[09:0010:00)
[10:0012:00)
[12:0014:00)
[14:0016:00)
[16:0017:00)
[17:0018:00)
[18:0019:00)
[19:0020:00)
[20:0022:00)
[22:0024:00)

Table 5
Comparison of level forecasting.

Method

Groupa

Mean MAE (veh/h/ln)

Group

Mean MAPE (%)

Group

Mean RMSE (veh/h/ln)

BATCH
AKF
KF
EXPRW

B
B
BA
A

43.5
44.98
46.44
50

B
B
B
A

7.15
7.31
7.69
8.4

B
BA
BA
A

65.91
68.78
69.63
74.78

Same code in the group column indicates no statistically signicant difference can be found in the mean performances of corresponding approaches.

tern, the random walk model part of EXPRW is overly simplistic for capturing the local variations in general. Second, AKF
outperforms KF consistently, indicating an increased performance of AKF over KF in terms of level forecasting. This demonstrates that improved performance can be expected through updating in real time the process variances for the Kalman
lters.
In addition to above overall performances, detailed level forecasting performances are shown in Figs. 1 and 2 with respect
to trafc levels and time of day, respectively. From these two gures, rst, for almost all trafc level and time of day groups,
the BARCH approach generates the best level forecasting performances and KF and AKF generate comparable performances
with AKF outperforms KF slightly. In addition, the EXPRW approach in general yields the worst performance except for trafc
level L5 when EXPRW shows the best performance. Note that for Level 5, trafc is near capacity with ow rate over 2000 veh/
h/ln and this high level trafc might not leave much room for the trafc oscillating. Therefore, simple forecasting approaches
such as EXPRW might be able to extrapolate trafc with desirable performance.
In summary, for level forecasting, though BATCH works best that agrees with previous ndings (e.g., Williams and Hoel,
2003; Smith et al., 2003), both KF and AKF perform reasonably well and AKF slightly outperforms KF. Considering the online
processing nature of AKF, it is fair to claim that AKF is promising for short term trafc ow level forecasting.

4.4. Results on uncertainty quantication

The results in terms of kickoff percentage with respect to trafc levels and time of day are presented in Figs. 3 and 4,
respectively. Note that these gures present the difference between the computed kickoff percentage and 5% for clarity purpose. First, it can be seen that in terms of all observations (for TTL in Fig. 3), though both KF and AKF are conservative compared to BATCH for generating prediction intervals at 95% signicant level, AKF outperforms KF in yielding kickoff percentage
closer to 5%; in addition, by further looking into the performance of AKF and KF for different trafc levels and time of day
groups, we can see that for all groups (G1 to G5) in Fig. 3 and most groups except for T3 and T4 in Fig. 4, AKF outperforms
KF, indicating the added adaptive ability integrated into AKF through estimating and updating recursively its process variances. Second, when investigating into the kickoff percentages of AKF and BATCH, we nd that for volatile trafc condition at
higher trafc levels (G3, G4, and G5 in Fig. 3) or for morning/afternoon rush hours (T4, T5, T6, T10, T11,T12, and T13 in Fig. 4),
AKF outperforms BATCH by yielding kickoff percentage closer to 5%. On reection, this nding might not be surprising in that
a xed set of parameters for BATCH is used for all trafc conditions without addressing trafc pattern variations, while AKF
can adjust the parameters with respect to trafc variations. This is desirable since most trafc management and control
operations are targeting volatile trafc conditions.

J. Guo et al. / Transportation Research Part C 43 (2014) 5064

59

Mean MAE (veh/hr/ln)

100
80
60

AKF
BATCH
EXPRW
KF

40
20

(a)

0
G1

G2

G3

G4

G5

Traffic Level
14

Mean MAPE (%)

12

AKF
BATCH
EXPRW
KF

10
8
6
4

(b)

2
0

G1

G2

G3

G4

G5

Traffic Level

Mean RMSE(veh/hr/ln)

140
120
100

AKF
BATCH
EXPRW
KF

80
60
40

(c)

20
0
G1

G2

G3

G4

G5

Traffic Level
Fig. 1. Level prediction performance comparison by trafc level.

The results of width to ow ratio with respect to trafc levels and time of day are presented in Figs. 5 and 6, respectively.
On observing the results, rst, we nd BATCH consistently generates smaller width to ow ratio than KF and AKF for all trafc levels and time of day; however, considering that BATCH has higher kickoff percentages for volatile trafc conditions, i.e.,
G3, G4, G5 in Fig. 3 and T4, T5, T6, T10, T11, T12, T13 in Fig. 4, the added value of narrower prediction intervals is offset by the
added kickoff percentages for these volatile trafc conditions. In addition, by comparing KF and AKF, it can be found that KF
outperforms AKF for all trafc levels and time of day groups; however, for volatile trafc conditions, the mean values of
width to ow ratios for AKF varies from 0.34 to 0.59, which is workable for these trafc conditions.
In summary, unlike the performances for level forecasting, the uncertainty quantication performances are mixed for
these approaches. However, considering the requirements of trafc management and control systems on volatile trafc conditions, it is safe to conclude that the online approach of AKF can generate workable prediction intervals for these volatile
trafc conditions.
4.5. Sensitivity analysis on AKF memory
In the AKF approach, the memory size N is critical for estimating the process variances that are to be adjusted into the
adaptive Kalman lter structure. For investigating the performances of AKF with respect to memory size, a sensitivity analysis of AKF memory size is conducted with results illustrated in Figs. 7 and 8 for MAPE and kickoff percentage, respectively.
For clarity, eight of the total 36 trafc ow series are selected. On observing the gures, we nd that AKF performances demonstrate a converging pattern when the memory size increases. For MAPE, the performance stabilizes when the memory size
is around 3 days, and for kickoff percentage, the performance stabilizes when the memory size is around 5 days. Therefore,
for simplicity and uniformity, the memory size is selected to 7 days (a week) in this paper, translating into 672 data point for
15-min interval. Considering this requirement, the eld implementation of AKF will need an adjusting time of about one
week to function normally. Taking advantage of the continuous nature of trafc data archiving systems, AKF can be initialized using historical trafc ow data for seeding the adaptive Kalman recursion.

J. Guo et al. / Transportation Research Part C 43 (2014) 5064

Mean MAE (veh/hr/ln)

60

100
80

(a)

60
AKF
BATCH
EXPRW
KF

40
20
0

T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 T11 T12 T13 T14 T15

Time of Day

Mean MAPE (%)

18
16
14
12
10
8
6
4

(b)

AKF
BATCH
EXPRW
KF

T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 T11 T12 T13 T14 T15

Mean RMSE (veh/hr/ln)

Time of Day
120
100

(c)

80
AKF
BATCH
EXPRW
KF

60
40
20
0

T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 T11 T12 T13 T14 T15

Time of Day
Fig. 2. Level prediction performance comparison by time of day.

10
9
8
7
6
5
4
3
2
1
0
TTL

BATCH
AKF
KF

G1

G2

G3

G4

G5

Traffic Level
Fig. 3. Kickoff percentage comparison by trafc level.

10
9
8
7
6
5
4
3
2
1
0

BATCH
AKF

KF

T1

T2

T3

T4

T5

T6

T7

T8

T9 T10 T11 T12 T13 T14 T15

Time of Day
Fig. 4. Kickoff percentage comparison by time of day.

61

Ratio

J. Guo et al. / Transportation Research Part C 43 (2014) 5064

2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0

BATCH
AKF
KF

TTL

G1

G2

G3

G4

G5

Traffic Level

Ratio

Fig. 5. Width to ow ratio comparison by trafc level.

2.4
2.2
2
1.8
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0

AKF
KF
BATCH

T1 T2 T3 T4 T5 T6 T7

T8 T9 T10 T11 T12 T13 T14 T15

Time of Day
Fig. 6. Width to ow ratio comparison by time of day.

20
UK 1996 4762a
UK 2002 6951a
MD 2004 9a

MAPE (%)

16

MD 2004 4b
MN 2000 442
MN 2004 737

12

WA 2004 ES_130D_MN_Stn
WA 2004 ES_738D_MN_Stn

4
1

AKF Memory size (day)

Fig. 7. Sensitivity of MAPE with respect to N.

5. Summaries and discussions

Over the decades, short term trafc ow forecasting has received a sustained attention from many transportation engineers and researchers for imparting the anticipatory ability required for developing proactive trafc control and management systems. Short term trafc ow forecasting includes predicting the level of trafc ow series, i.e., to predict how
many vehicles will be arriving within the next time interval, and constructing the prediction interval, i.e., to decide the dispersion (uncertainty) of the prediction. Recently, studies shows that the conditional mean and the conditional variance of
trafc ow series can be modeled as the SARIMA + GARCH structure. Based on this structure, online algorithm can be developed for processing this structure consecutively through a seasonal exponential smoothing operator and two state space representation based Kalman ltering models. However, for conventional Kalman lters, the seeding process variances are
constants that are estimated from historical trafc ow data. Therefore, in this paper, the adaptive Kalman recursion is investigated to estimate and update the process variances in real time.
The proposed adaptive Kalman lters are applied for level forecasting and prediction interval construction, respectively.
Using real world data from four regions around the world, the adaptive Kalman lters are compared with three other approaches. For level forecasting, the empirical results show the added performance of adaptive Kalman lter over conven-

J. Guo et al. / Transportation Research Part C 43 (2014) 5064

Prediction interval kickoff percentage (%)

62

15
UK 1996 4762a
UK 2002 6951a
MD 2004 9a
MD 2004 4b

10

MN 2000 442
MN 2004 737
WA 2004 ES_130D_MN_Stn
WA 2004 ES_738D_MN_Stn

AKF memory size (day)

Fig. 8. Sensitivity of kickoff percentage with respect to N.

tional Kalman lter through updating the process variances. For uncertainty quantication, the results are mixed while the
empirical investigation shows that the adaptive Kalman lter can generate workable prediction intervals; in particular, the
adaptive Kalman lter approach showed improved adaptability when trafc is highly volatile. Sensitivity analysis is also conducted, showing that the performance of the adaptive Kalman lter stabilizes with the increase of its memory size.
Some remarks are provided as follows. First, it should be noted that the performances of AKF are affected fundamentally
by the stochastic structure from which the state space models are developed. Therefore, compared with level forecasting
where SARIMA has been shown in many studies to be able to capture the rst conditional moment of trafc ow series,
GARCH model is only adopted for uncertainty quantication in the past few years. Consequently, it is not surprising to
see the mixed results in uncertainty quantication performances. In fact, future studies should be conducted to advance
the investigation into the second order moment of trafc condition series.
Second, uniform performance measures should be investigated to evaluate the prediction intervals. In this paper, kickoff
percentage and width to ow level are used to gauge the coverage and length of the prediction intervals, respectively.
However, a uniform measure is still lacking to determine the overall performance. Therefore, more work should be conducted to evaluate the performance of prediction intervals, in particular, for cases when the coverage measure conicts with
the length measure as found in this paper.
In the end, more adaptation mechanism should be investigated. As discussed in Noriega and Pasupathy (1997), the process noise adaptation mechanism adopted in this study is a suboptimal sequential estimator which can handle time variant
models with relatively modest computational cost. However, for more improvements, further investigations are recommended on the variations of adaptation mechanisms.
Acknowledgements
The authors would like to thank the Minnesota Department of Transportation, the Washington State Department of
Transportation, the Maryland Department of Transportation, and the United Kingdoms Highways Agency for providing
the data used in this study. This work was supported in part by the Natural Science Foundation of China under Grant
Nos. 71101025 and 60904026, the National Key Technology R&D Program under grant No. 2011BAK21B01, and the Fundamental Research Funds for the Central Universities.
References
Ahmed, M.S., Cook, A.R., 1979. Analysis of freeway trafc time-series data by using Box-Jenkins techniques. Transp. Res. Rec. 722, 19.
Bollerslev, T., 1986. Generalized autoregressive conditional heteroscedasticity. J. Econ. 31, 307327.
Box, G.E.P., Jenkins, G.M., Reinsel, G.C., 1994. Time Series Analysis: Forecasting and Control, third ed. Prentice-Hall Inc., Englewood Cliff, New Jersey.
Box, G.E.P., Jenkins, G.M., Reinsel, G.C., 2008. Time Series Analysis: Forecasting and Control, fourth ed. John Wiley & Sons Inc., Hoboken, New Jersey.
Chateld, C., 1993. Calculating interval forecasts. J. Business Econ. Stat. 11 (2), 121135.
Chen, H., Grant-Muller, S., 2001. Use of sequential learning for short-term trafc ow forecasting. Transp. Res. Part C 9, 319336.
Chen, C., Hu, J., Meng, Q., Zhang, Y., 2011. Short-time trafc ow prediction with ARIMA GARCH model. Intelligent Vehicles Symposium (IV), IEEE, pp. 607
612.
Clark, S.D., Dougherty, M.S., Kirby, H.R., 1992. Statistical Models for Predicting Short-term Changes in Trafc Flows: A Comparison with Neural Networks. ITS
Technical Note 317, Institute for Transport Studies, University of Leeds, United Kingdom.
Danech-Pajouh, M., Aron, M., 1991. ATHENA A Method for Short-term Inter-urban Trafc Forecasting. English Version of Report 177, INRETS, Arcueil, Paris.
Davis, G.A., Nihan, N.L., 1991. Nonparametric regression and short-term freeway trafc forecasting. ASCE J. Transp. Eng. 117, 178188.
Der Voort, M., Dougherty, M., Watson, S., 1996. Combining Kohonen maps with ARIMA time series models to forecast trafc ow. Transp. Res. Part C 4, 307
318.

J. Guo et al. / Transportation Research Part C 43 (2014) 5064

63

Dia, H., 2001. An object-oriented neural network approach to short-term trafc forecasting. Eur. J. Oper. Res. 131, 253261.
Dimitriou, L., Tsekeris, T., Stathopoulos, A., 2008. Adaptive hybrid fuzzy rule-based system approach for modeling and predicting urban trafc ow. Transp.
Res. Part C 16, 554573.
Dunne, S., Ghosh, B., 2012. Regime-based short-term multivariate trafc condition forecasting algorithm. ASCE J. Transp. Eng. 138, 455466.
Edie, L.C., 1963. Discussion on trafc stream measurements and denitions. In: Proceedings of 2nd International Symposium on the Theory of Trafc Flow,
OECD, Paris, France, pp. 139154.
El Faouzi, N., 1996. Nonparametric trafc ow prediction using kernel estimation. In: Proceedings of the 13th International Symposium on Transportation
and Trafc Theory, pp. 4154.
Engle, R.F., 1982. Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom ination. Econometrica 50, 9871008.
Fei, X., Lu, C., Liu, K., 2011. A Bayesian dynamic linear model for real-time short-term freeway travel time prediction. Transp. Res. Part C 19, 13061318.
Fuller, W.A., 1996. Introduction to Statistical Time Series, second ed. John Wiley & Sons, Hoboken, New Jersey.
Gazis, D.C., Knapp, C.H., 1971. On-line estimation of trafc densities from time-series of ow and speed data. Transp. Sci. 5, 283301.
Guo, J., 2005. Adaptive Estimation and Prediction of Univariate Trafc Condition Series. Ph.D. Dissertation. North Carolina State University, Raleigh, NC.
Guo, J., Williams, B.M., 2010. Real time short term trafc speed level forecasting and uncertainty quantication using layered Kalman lters. Transp. Res.
Rec. 2175, 2837.
Guo, J., Williams, B.M., Smith, B.L., 2008. Data collection time intervals for stochastic short-term trafc ow forecasting. Transp. Res. Rec. 2024, 1826.
Guo, J., Huang, W., Williams, B.M., 2012. Integrated heteroscedasticity test for vehicular trafc condition series. ASCE J. Transp. Eng. 138, 11611170.
Hamed, M.M., Al-Masaeid, H.R., Said, Z.M.B., 1995. Short-term prediction of trafc volume in urban arterials. ASCE J. Transp. Eng. 12, 249254.
Hugosson, M.B., 2005. Quantifying uncertainties in a national forecasting model. Transp. Res. Part A 39, 531547.
Jiang, X., Adeli, H., 2005. Dynamic wavelet neural network model for trafc ow forecasting. ASCE J. Transp. Eng. 131, 771779.
Kalman, R.E., 1960. A new approach to linear ltering and prediction problems. Trans. ASME J. Basic Eng. D 82, 3545.
Kamarianakis, Y., Prastacos, P., 2003. Forecasting trafc ow conditions in an urban network: comparison of multivariate and univariate approaches. Transp.
Res. Rec. 1857, 7484.
Kamarianakis, Y., Kanas, A., Prastacos, P., 2005. Modeling trafc volatility dynamics in an urban network. Transp. Res. Rec. 1923, 1827.
Kang, S., Ritchie, S.G., Jayakrishnan, R., 1998. Prediction of short-term freeway trafc volume using recursive least square and lattice ltering. In:
Proceedings of the 5th International Conference on Applications of Advanced Technologies in Transportation Engineering, pp. 255264.
Karlaftis, M.G., Vlahogianni, E.I., 2009. Memory properties and fractional integration in transportation time series. Transp. Res. Part C 17, 444453.
Khosravi, A., Mazloumi, E., Nahavandi, S., Creighton, D., Van Lint, J., 2011. A genetic algorithm-based method for improving quality of travel time prediction
intervals. Transp. Res. Part C: Emerg. Technol. 19, 13641376.
Lan, C., 2001. A recursive trafc ow predictor based on dynamic generalized linear model framework. In: IEEE Intelligent Transportation Systems
Proceedings, Oakland, CA, pp.410415.
Lan, C., Miaou, S., 1999. Real-time prediction of trafc ows using dynamic generalized linear models. Transp. Res. Rec. 1687, 410415.
Levin, M., Tsao, Y., 1980. On forecasting freeway occupancies and volumes. Transp. Res. Rec. 773, 4749.
Lippi, M., Bertini, M., Frasconi, P., 2013. Short-term trafc ow forecasting: an experimental comparison of time series analysis and supervised learning. IEEE
Trans. Intell. Transp. Syst. 14, 871882.
Lu, J., 1990. Prediction of trafc ow by an adaptive prediction system. Transp. Res. Rec. 1287, 5461.
Mazloumi, E., Rose, G., Currie, G., Moridpour, S., 2011. Prediction intervals to account for uncertainties in neural network predictions: methodology and
application in bus arrival time prediction. Eng. Appl. Artif. Intell. 24, 534542.
Min, W., Wynter, L., 2011. Real-time road trafc prediction with spatio-temporal correlations. Transp. Res. Part C 19, 606616.
Myers, K.A., Tapley, B.D., 1976. Adaptive sequential estimation with unknown noise structure. IEEE Trans. Autom. Control 21, 520523.
Nicholson, H., Swann, C.D., 1974. The prediction of trafc ow volume based on spectral analysis. Transp. Res. 8, 533538.
Nihan, N.L., Holersland, K.O., 1980. Use of the Box-Jenkins time series technique in trafc forecasting. Transportation 9, 125143.
Noriega, G., Pasupathy, S., 1997. Adaptive estimation of noise covariance matrices in real-time preprocessing of geophysical data. IEEE Trans. Geosci. Remote
Sens. 35, 11461159.
Okutani, I., Stephanedes, Y.J., 1984. Dynamic prediction of trafc volume through Kalman ltering theory. Transp. Res. Part B 18, 111.
Park, B., Messer, C.J., Urbanik, T., 1998. Short-term freeway trafc volume forecasting using radial basis function neural network. Transp. Res. Rec. 1651, 39
47.
Pattanamekar, P., Dongjoo, P., Laurence, R.R., Lee, J., Lee, C., 2003. Dynamic and stochastic shortest path in transportation networks with two components of
travel time uncertainty. Transp. Res. Part C 11, 331354.
Ross, P., 1982. Exponential ltering of trafc data. Transp. Res. Rec. 869, 4349.
SAS Institute Inc., 1999. SAS OnlineDoc, Version 8. SAS Institute Inc., Cary, NC
Shekhar, S., 2004. Recursive Methods for Forecasting Short-term Trafc Flow using Seasonal ARIMA Time Series Model. Master Thesis. North Carolina State
University, Raleigh, NC.
Shekhar, S., Williams, B.M., 2008. Adaptive seasonal time series models for forecasting short-term trafc ow. Transp. Res. Rec. 2024, 116125.
Smith, B.L., Demetsky, M.J., 1994. Short-term trafc ow prediction: neural network approach. Transp. Res. Rec. 1453, 98104.
Smith, B.L., Demetsky, M.J., 1996. Multiple-interval freeway trafc ow forecasting. Transp. Res. Rec. 1554, 136141.
Smith, B.L., Demetsky, M.J., 1997. Trafc ow forecasting: comparison of modeling approaches. ASCE J. Transp. Eng. 123, 261266.
Smith, B.L., Williams, B.M., Oswald, R.K., 2003. Comparison of parametric and non-parametric models for trafc ow forecasting. Transp. Res. Part C 10, 303
321.
Sohn, K., Kim, D., 2009. Statistical model for forecasting link travel time variability. ASCE J. Transp. Eng. 135 (7), 440453.
Stathopoulos, A., Karlaftis, M.G., 2003. A multivariate state space approach for urban trafc ow modeling and prediction. Transp. Res. Part C 11, 121135.
Stephanedes, Y.J., Michalopoulos, P.G., Plum, R.A., 1981. Improved estimation of trafc ow for real-time control. Transp. Res. Rec. 795, 2839.
Sun, H., Liu, H.X., Xiao, H., He, R.R., Ran, B., 2003. Short-term trafc forecasting using the local linear regression model. Transp. Res. Rec. 1836, 143150.
Sun, H., Zhang, C., Ran, B., Choi, K., 2004. Prediction Interval for Trafc Time Series. 83rd TRB CD Washington, DC.
Tchrakian, T.T., Basu, B., OMahony, M., 2012. Real-time trafc ow forecasting using spectral analysis. IEEE Trans. Intell. Transp. Syst. 13 (2), 519526.
TRB, 2000. Highway Capacity Manual, Transportation Research Board, Washington, DC.
Tsekeris, T., Stathopoulos, A., 2006. Real-time trafc volatility forecasting in urban arterial networks. Transp. Res. Rec. 1964, 146156.
Tsekeris, T., Stathopoulos, A., 2010. Short-term prediction of urban trafc variability: stochastic volatility modeling approach. ASCE J. Transp. Eng. 136 (7),
606613.
Van Hinsbergen, C.P.I.J., van Lint, J.W.C., van Zuylen, H.J., 2009. Bayesian committee of neural networks to predict travel times with condence intervals.
Transp. Res. Part C 17, 498509.
Vlahogianni, E.I., Karlaftis, M.G., Golias, J.C., 2005. Optimized and meta-optimized neural networks for short-term trafc ow prediction: a genetic approach.
Transp. Res. Part C 13, 211234.
Wang, J., Shi, Q., 2013. Short-term trafc speed forecasting hybrid model based on chaos-wavelet analysis-support vector machine theory. Transp. Res. Part
C 27, 219232.
Wei, Y., Chen, M., 2012. Forecasting the short-term metro passenger ow with empirical mode decomposition and neural networks. Transp. Res. Part C 21,
148162.
Williams, B.M., 1999. Modeling and Forecasting Vehicular Trafc Flow as a Seasonal Stochastic Time Series Process. Ph.D. Dissertation. Dept. of Civil Eng.,
Univ. of Virginia, Charlottesville, VA.

64

J. Guo et al. / Transportation Research Part C 43 (2014) 5064

Williams, B.M., 2001. Multivariate vehicular trafc ow prediction: evaluation of ARIMAX modeling. Transp. Res. Rec. 1776, 194200.
Williams, B.M., Hoel, L.A., 2003. Modeling and forecasting vehicular trafc ow as a seasonal ARIMA: theoretical basis and empirical results. ASCE J. Transp.
Eng. 129, 664672.
Yang, F., Yin, Z., Liu, H.X., Ran, B., 2004. Online recursive algorithm for short-term trafc prediction. Transp. Res. Rec. 1879, 18.
Yang, M., Liu, Y., You, Z., 2010. The reliability of travel time forecasting. IEEE Trans. Intell. Transp. Syst. 11 (1), 162171.
Yin, H., Wong, S.C., Xu, J., Wong, C.K., 2002. Urban trafc ow prediction using a fuzzy-neural approach. Transp. Res. Part C 10, 8598.
Zhang, H.M., 2000. Recursive prediction of trafc conditions with neural network models. ASCE J. Transp. Eng. 126, 472481.
Zhang, Y., Sun, R., Haghani, A., Zeng, X., 2013. Univariate volatility-based models for improving quality of travel time reliability forecasting. Transp. Res. Rec.
2365, 7381.
Zheng, W., Lee, D., Shi, Q., 2006. Short-term freeway trafc ow prediction: Bayesian combined neural network approach. ASCE J. Transp. Eng. 132, 114121.