Structural Identification Scheme Research Articles

State estimation and structure identification of complex networks with time‐varying delays and model uncertainty based on probabilistic graph neural network

SummaryIn this paper, the state estimation problem and structure identification problem are investigated for an array of complex networks with time‐varying delays and model uncertainty. Model uncertainty is generated by uncertain connections between nodes in complex networks. Combining probabilistic graph model and graph neural network, a probabilistic graph neural network is proposed to deal with the uncertainty, and the multiple linear terms of the errors between nodes are estimated. The purpose of the addressed state estimation problem is to design a state estimator such that, in the presence of the model uncertainty, the estimation error to converge to zero can be guaranteed and the explicit expression of the estimator parameters is given. Based on a Lyapunov function, the parameters of the estimator and the updating laws of the weights of probabilistic graph neural network can be obtained by the solutions to linear matrix inequalities. The purpose of the addressed structure identification problem is to infer a definite graph structure model based on Bayes' theorem in the presence of a stable probabilistic graph neural network and state estimator. Finally, an illustrative example is provided to demonstrate the feasibility and effectiveness of the developed state estimation and structure identification scheme.

EVALUATION OF MONETARY POLICY SHOCKS IN TURKEY: A STRUCTURAL VAR APPROACH

This paper empirically investigates the effects of monetary policy shocks on the Turkish economy using a structural VAR model. Monetary policy shocks are identified based on the non-recursive structural identification scheme. Since monetary policy stance is contingent on different funding rates in the wide interest rate corridor for the selected sample period, 2011:M1-2018:M12, this paper employs “weighted average funding cost”to represent the monetary policy stance of the CBRT. The baseline identification scheme is also extended in different ways to check the robustness of the results. The empirical results can be summarized as follows. First, Turkish data are not free from price and exchange rate puzzles. More importantly, qualitative inferences are quite persistent across different identification restrictions. Second, structural impulse-responses reveal that there is a two-way simultaneous interaction between monetary policy and exchange rate. Finally, the presence of money stock in the VAR model is redundant for identifying monetary policy shocks since the VAR models with and without money stock generate almost identical results.

The credit supply channel of monetary policy: evidence from a FAVAR model with sign restrictions

We test whether the credit channel of the monetary policy was present in the United States’ economy from January 2001 to April 2016. To this end, we use a factor-augmented vector autoregression, and we impose sensible theoretical sign restrictions in our structural identification scheme. We use the expected substitution effect between bank commercial loans and commercial papers to identify the credit supply channel. We found that the credit channel appears to have operated in the US economy during the sample period. However, when we split the sample, we found that the credit channel did not operate after the subprime crisis (close to the Zero Lower Bound of the interest rate). This result is robust to changing the sign restriction horizons. It supports current views in the literature regarding the ineffectiveness of the credit channel as a means to foster real economic activity during crises episodes.

On the importance of testing structural identification schemes and the potential consequences of incorrectly identified models

Identification schemes are of essential importance in structural analysis. This paper focuseson testing a commonly used long-run structural parameter identification scheme claiming to identifyfundamental and non-fundamental shocks to stock prices. Five related widely used structural modelson assessing stock price determinants are considered. All models are either specified in vector errorcorrection (VEC) or in vector autoregressive (VAR) form. A Markov switching in heteroskedasticitymodel is used to test the identifying restrictions. It is found that for two of the models considered,the long-run identification scheme appropriately classifies shocks as being either fundamental or nonfundamental.A small empirical exercise finds that the models with properly identified structuralshocks deliver realistic conclusions, similar as in some of the literature. On the other hand, modelswith identification schemes not supported by the data yield dubious conclusions on the importance offundamentals for real stock prices. This is because their structural shocks are not properly identified,making any shock labelling ambiguous. Hence, in order to ensure that economic shocks of interest areproperly captured, it is important to test the structural identification scheme.

On the importance of testing structural identification schemes and the potential consequences of incorrectly identified models

Identification schemes are of essential importance in structural analysis. This paper focuses on testing a commonly used long-run structural parameter identification scheme claiming to identify fundamental and non-fundamental shocks to stock prices. Five related widely used structural models on assessing stock price determinants are considered. All models are either specified in vector error correction (VEC) or in vector autoregressive (VAR) form. A Markov switching in heteroskedasticity model is used to test the identifying restrictions. It is found that for two of the models considered, the long-run identification scheme appropriately classifies shocks as being either fundamental or non-fundamental. A small empirical exercise finds that the models with properly identified structural shocks deliver realistic conclusions, similar as in some of the literature. On the other hand, models with identification schemes not supported by the data yield dubious conclusions on the importance of fundamentals for real stock prices. This is because their structural shocks are not properly identified, making any shock labelling ambiguous. Hence, in order to ensure that economic shocks of interest are properly captured, it is important to test the structural identification scheme.

A visual assessment of hairpin packet structures in a DNS of a turbulent boundary layer

A well-resolved DNS of a zero-pressure-gradient turbulent boundary layer at moderately high Reynolds number is analyzed in order to visually assess, in three dimensions, the existence of instantaneous realizations of outer region hairpins and hairpin packets. For this purpose, fifty independent 2D streamwise-wall-normal slices of DNS data at Reθ=4300 are employed to mimic lower-resolution PIV velocity fields using the mimicking procedure of Rahgozar et al. (2013). The two-dimensional structure identification scheme of Adrian et al. (2000) is then applied on this resolution-degraded DNS data in order to discover the signatures of hairpins and hairpin packets. Subsequently, the 2D motions that are identified as signatures of hairpins and hairpin packets are isolated and visualized in three dimensions using the fully resolved DNS data. In agreement with Adrian et al. (2000), the hairpin and hairpin packet signatures are frequently observed in the mimicked planes. However, the 3D character of the 2D signatures is found to be significantly different from the inferred hairpins and hairpin packets. The 2D packet signatures, when viewed in 3D, generally correspond to a series of irregular and complex vortex tubes which are not necessarily different from other nearby vortex tubes in terms of their shape, size and intensity.

A new structure identification scheme for ANFIS and its application for the simulation of virtual air pollution monitoring stations in urban areas

Parameter and structure identifications are necessary in any modelling which aims to achieve a generalised model. Although ANFIS (Adaptive Network-based Fuzzy Inference System) employs well-known parameter-identification techniques, it needs to structure identification techniques for the determination of an optimum number of fuzzy rules and the selection of significant input variables from among the candidate input variables. In this study, a new structure identification scheme is developed and introduced, which is simultaneously capable of the selection of significant input variables and the determination of an optimum number of rules. This new structure identification was joined to ANFIS, and this joined modelling framework was applied to the simulation of virtual air-pollution monitoring stations in Berlin. In this study, 18 virtual particulate matter stations were simulated using the particulate matter data of some of the current stations. In other words, the particulate matter monitoring network of Berlin has been intensified. The evaluation of simulated virtual stations shows that, although the uncertainty of daily particulate matter measurement is about 10 percent, the simulated virtual stations can estimate the mean daily particulate matter with less than 10 percent of error. Mean absolute error and root mean square error of the simulations are less than 2.4 and 3.4µg/m3, respectively. The correlation coefficient of the simulation results was more than 0.94. In addition, the range of mean bias error is between −1.0 and 0.5µg/m3, and the range of factor of exceedance is between −14.8 and 10.8 percent. It means that the simulated virtual stations have a small bias. These results demonstrated the appropriate performance of the joined new structure identification scheme and ANFIS for development of a virtual air pollution monitoring network.

Can Stock Price Fundamentals Properly Be Captured? Using Markov Switching in Heteroskedasticity Models to Test Identification Schemes

Structural identification schemes are of essential importance to vector autoregressive (VAR) analysis. This paper tests a commonly used structural parameter identification scheme to assess whether it can properly capture fundamental and non-fundamental shocks to stock prices. In particular, five related structural models, which are widely used in the literature on assessing stock price determinants are considered. They are either specified in vector error correction (VEC) or in VAR form. Restrictions on the long-run effects matrix are used to identify the structural parameters. These identifying restrictions are tested by means of a Markov switching in heteroskedasticity model. It is found that for two of the five models considered, the long-run identification scheme appropriately classifies shocks as being either fundamental or non-fundamental. A series of robustness tests are performed, which largely confirm the initial findings.

An on-line ICA-mixture-model-based self-constructing fuzzy neural network

This paper proposes a new fuzzy neural network (FNN) capable of parameter self-adapting and structure self-constructing to acquire a small number of fuzzy rules for interpreting the embedded knowledge of a system from the given training data set. The proposed FNN is inherently a modified Takagi-Sugeno-Kang (TSK)-type fuzzy-rule-based model with neural network's learning ability. There are no rules initiated at the beginning and they are created and adapted through an on-line learning processing that performs simultaneous structure and parameter identification. In the structure identification of the precondition part, the input space is partitioned in a flexible way according to the newly proposed on-line independent component analysis (ICA) mixture model. The input space is thus represented by linear combinations of independent, non-Gaussian densities. The first input training pattern is assigned to the first rule initially by the on-line ICA mixture model. Afterwards, some additional significant terms (input variables) selected by the on-line ICA mixture model will be added to the consequent part (forming a liner equation of input variables) incrementally or create a new rule in the learning processing. The combined precondition and consequent structure identification scheme can make the network grow dynamically and efficiently. In the parameter identification, the consequent parameters are tuned by the backpropagation rule and the precondition parameters are turned by the on-line ICA mixture model. Both the structure and parameter identifications are done simultaneously to form a fast learning scheme. The derived on-line ICA mixture model also provide a natural linear transformation for each input variable to enhance the knowledge representation ability of the proposed FNN and reduce the required rules and achieve higher accuracy efficiently. In order to demonstrate the performance of the proposed FNN, several experiments covering the areas of system identification, classification, and image segmentation are carried out. Our experiments show that the proposed FNN can achieve significant improvements in the convergence speed and prediction accuracy with simpler network structure.

Applications of sliding mode control to benchmark problems

In this paper, both the methods of continuous sliding mode control (CSMC) and continuous sliding mode control with compensators (CSMC&C) have been applied to two benchmark structures, namely, a building model equipped with an active mass driver system, and a building model equipped with an active tendon system. The CSMC&C strategy is a modification of CSMC to facilitate the design of static output feedback controllers and to provide a systematic tuning of the control effort. Due to the structural identification scheme used in the benchmark problems, in which the state variables are fictitious, one cannot take the full advantages of static output feedback controllers. As a result, an observer is used in CSMC, whereas a low-pass filter is incorporated for each measurement in CSMC&C. The purpose of using low-pass filters in CSMC&C is to transform the benchmark problems into strictly proper systems. The main advantage of the CSMC&C method is that the on-line computational effort is reduced since the dimension of filters and compensator is much smaller than that of an observer. Simulation results based on the CSMC and CSMC&C methods are presented and compared with that of the LQG method. Robustness of stability and noise rejection for each controller design are also illustrated by examining the loop transfer function. Simulation results for the benchmark problems indicate that the control performances for LQG, CSMC and CSMC&C are quite comparable.

An online self-constructing neural fuzzy inference network and its applications

A self-constructing neural fuzzy inference network (SONFIN) with online learning ability is proposed in this paper. The SONFIN is inherently a modified Takagi-Sugeno-Kang (TSK)-type fuzzy rule-based model possessing neural network learning ability. There are no rules initially in the SONFIN. They are created and adapted as online learning proceeds via simultaneous structure and parameter identification. In the structure identification of the precondition part, the input space is partitioned in a flexible way according to an aligned clustering-based algorithm. As to the structure identification of the consequent part, only a singleton value selected by a clustering method is assigned to each rule initially. Afterwards, some additional significant terms selected via a projection-based correlation measure for each rule will be added to the consequent part incrementally as learning proceeds. The combined precondition and consequent structure identification scheme can set up an economic and dynamically growing network, a main feature of the SONFIN. In the parameter identification, the consequent parameters are tuned optimally by either least mean squares or recursive least squares algorithms and the precondition parameters are tuned by a backpropagation algorithm. To enhance the knowledge representation ability of the SONFIN, a linear transformation for each input variable can be incorporated into the network so that much fewer rules are needed or higher accuracy can be achieved.

Parameter and structure identification of linear multivariable systems

In this paper, a parameter and structure identification scheme, which can be used for both continuous and discrete time systems, is presented. This scheme has an “on-line” nature and it incorporates standard techniques such as state variable filters and adaptation laws developed through stability analysis. The canonical form being used is based on the output injection which is constructed to assign all the poles of the system to α, an arbitrary chosen real number. It is therefore named the “α-canonical form”.