A new traffic forecasting method using state space representation is proposed. By means of a state space model, the forecasting value is sequentially calculated by applying the Kalman filter. However the true traffic structure is not easy to grasp as changes in traffic are largely due to social activities. In addition, experience has shown that economic trends in society also have an influence on traffic. For this reason, the traffic structure becomes too complex to describe changes in traffic by using a single state space model. In this paper a multiple state space model is proposed. The multiple state space models is composed of several state space models calls sub-models. This model is more easily adaptable to change in the traffic structure than a single state space model. The Bayesian forecasting value is given by the weighted summation for each sub-model forecasting value. The Bayesian posterior probability, which is calculated from the likelihood, is used as the weight of the sub-model. A good fitting sub-model posterior probability increases as the number of observations increases. In this paper the initial state and noise variances of each sub-model are estimated by numerical maximization of the likelihood. Examples of how this method may be applied to monthly telephone revenue data and trunk group load data are given, demonstrating the possibility of adapting exceptional data and structural changes in traffic. Parameter estimation using a multiple state space model is also shown.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Takeo ABE, Hiroshi SAITO, "Bayesian Forecasting with Multiple State Space Model" in IEICE TRANSACTIONS on transactions,
vol. E69-E, no. 3, pp. 210-216, March 1986, doi: .
Abstract: A new traffic forecasting method using state space representation is proposed. By means of a state space model, the forecasting value is sequentially calculated by applying the Kalman filter. However the true traffic structure is not easy to grasp as changes in traffic are largely due to social activities. In addition, experience has shown that economic trends in society also have an influence on traffic. For this reason, the traffic structure becomes too complex to describe changes in traffic by using a single state space model. In this paper a multiple state space model is proposed. The multiple state space models is composed of several state space models calls sub-models. This model is more easily adaptable to change in the traffic structure than a single state space model. The Bayesian forecasting value is given by the weighted summation for each sub-model forecasting value. The Bayesian posterior probability, which is calculated from the likelihood, is used as the weight of the sub-model. A good fitting sub-model posterior probability increases as the number of observations increases. In this paper the initial state and noise variances of each sub-model are estimated by numerical maximization of the likelihood. Examples of how this method may be applied to monthly telephone revenue data and trunk group load data are given, demonstrating the possibility of adapting exceptional data and structural changes in traffic. Parameter estimation using a multiple state space model is also shown.
URL: https://global.ieice.org/en_transactions/transactions/10.1587/e69-e_3_210/_p
Copy
@ARTICLE{e69-e_3_210,
author={Takeo ABE, Hiroshi SAITO, },
journal={IEICE TRANSACTIONS on transactions},
title={Bayesian Forecasting with Multiple State Space Model},
year={1986},
volume={E69-E},
number={3},
pages={210-216},
abstract={A new traffic forecasting method using state space representation is proposed. By means of a state space model, the forecasting value is sequentially calculated by applying the Kalman filter. However the true traffic structure is not easy to grasp as changes in traffic are largely due to social activities. In addition, experience has shown that economic trends in society also have an influence on traffic. For this reason, the traffic structure becomes too complex to describe changes in traffic by using a single state space model. In this paper a multiple state space model is proposed. The multiple state space models is composed of several state space models calls sub-models. This model is more easily adaptable to change in the traffic structure than a single state space model. The Bayesian forecasting value is given by the weighted summation for each sub-model forecasting value. The Bayesian posterior probability, which is calculated from the likelihood, is used as the weight of the sub-model. A good fitting sub-model posterior probability increases as the number of observations increases. In this paper the initial state and noise variances of each sub-model are estimated by numerical maximization of the likelihood. Examples of how this method may be applied to monthly telephone revenue data and trunk group load data are given, demonstrating the possibility of adapting exceptional data and structural changes in traffic. Parameter estimation using a multiple state space model is also shown.},
keywords={},
doi={},
ISSN={},
month={March},}
Copy
TY - JOUR
TI - Bayesian Forecasting with Multiple State Space Model
T2 - IEICE TRANSACTIONS on transactions
SP - 210
EP - 216
AU - Takeo ABE
AU - Hiroshi SAITO
PY - 1986
DO -
JO - IEICE TRANSACTIONS on transactions
SN -
VL - E69-E
IS - 3
JA - IEICE TRANSACTIONS on transactions
Y1 - March 1986
AB - A new traffic forecasting method using state space representation is proposed. By means of a state space model, the forecasting value is sequentially calculated by applying the Kalman filter. However the true traffic structure is not easy to grasp as changes in traffic are largely due to social activities. In addition, experience has shown that economic trends in society also have an influence on traffic. For this reason, the traffic structure becomes too complex to describe changes in traffic by using a single state space model. In this paper a multiple state space model is proposed. The multiple state space models is composed of several state space models calls sub-models. This model is more easily adaptable to change in the traffic structure than a single state space model. The Bayesian forecasting value is given by the weighted summation for each sub-model forecasting value. The Bayesian posterior probability, which is calculated from the likelihood, is used as the weight of the sub-model. A good fitting sub-model posterior probability increases as the number of observations increases. In this paper the initial state and noise variances of each sub-model are estimated by numerical maximization of the likelihood. Examples of how this method may be applied to monthly telephone revenue data and trunk group load data are given, demonstrating the possibility of adapting exceptional data and structural changes in traffic. Parameter estimation using a multiple state space model is also shown.
ER -