Handling missing data in multivariate time series using a vector autoregressive model based imputation (VAR-IM) algorithm: Part I: VAR-IM algorithm versus traditional methods

Vector autoregressive model proposed for multivariate time series data. Neural Network, including Feed Forward Neural Network (FFNN), is the powerful tool for the nonlinear model. In autoregressive model, the input layer is the past values of the same series up to certain lag and the output layers is the current value. So, VAR-NN is proposed to predict the multivariate time series data using nonlinear approach. The optimal lag time in VAR are used as aid of selecting the input in VAR-NN. In this study we develop the soft computation tools of VAR-NN based on Graphical User Interface. In each number of neurons in hidden layer, the looping process is performed several times in order to get the best result. The best one is chosen by the least of Mean Absolute Percentage Error (MAPE) criteria. In this study, the model is applied in the two series of stock price data from Indonesia Stock Exchange. Evaluation of VAR-NN performance was based on train-validation and test-validation sample approach. Based on the empirical stock price data it can be concluded that VAR-NN yields perfect performance both in in-sample and in out-sample for non-linear function approximation. This is indicated by the MAPE value that is less than 1% .

Download Full-text

Modelling Inflation Sectors in Indonesia Using Vector Autoregressive (VAR)

Jurnal ILMU DASAR ◽

10.19184/jid.v20i1.7259 ◽

2019 ◽

Vol 20 (1) ◽

pp. 47

Author(s):

Alan Prahutama ◽

S. Suparti ◽

Dwi Ispriyanti ◽

Tiani Wahyu Utami

Keyword(s):

Time Series ◽

White Noise ◽

Granger Causality ◽

Autoregressive Model ◽

Vector Autoregressive Model ◽

Vector Autoregressive

Analisis time series dapat dibagi menjadi dua yaitu analisis time series univariat dan analisis time series multivariat. Analisis time series univariat salah satunya menggunakan ARIMA, sedangkan analisis time series multivariat dapat menggunakan VAR. VAR merupakan pemodelan persamaan simultan yang memiliki beberapa variabel endogen secara bersamaan. Asumsi dalam model VAR antara lain terjadi kausalitas antar variabel (kausalitas Granger), residual white noise dan berdistribusi normal multivariat. Pada paper ini, metode VAR diimplementasikan dalam memodelkan sektor-sektor Inflasi di Indonesia. Adapun sektor-sektor tersebut antara lain sektor makanan (Y1t),Sektor Makanan Jadi, Minuman, Rokok dan Tembakau (Y2), Sektor perumahan, listrik, air, gas dan bahan bakar (Y3), Sektor Sandang (Y4), Sektor Kesehatan (Y5), Sektor Pendidikan dan Olahraga (Y6), Sektor Transportasi, Komunikasi dan Jasa Keuangan (Y7). Hasilnya adalah tidak semua variabel sektor inflasi berpengaruh terhadap sektor lainnya. Hanya beberapa variabel yang berpengaruh terhadap suatu sektor. Asumsi kausalitas Granger tidak semua dipenuhi oleh semua variabel. Begitu juga dengan normal multivariat juga tidak terpenuhi. Akan tetapi residual model sudah white noise. Keywords: vector autoregressive model, sectors of inflation, Granger Causality.

Download Full-text

Use of Vector Autoregressive Model for Anomaly Detection in Utility Gas Turbines

Volume 3: Coal, Biomass, Hydrogen, and Alternative Fuels; Cycle Innovations; Electric Power; Industrial and Cogeneration; Organic Rankine Cycle Power Systems ◽

10.1115/gt2019-90995 ◽

2019 ◽

Author(s):

Vipul Goyal ◽

Mengyu Xu ◽

Jayanta Kapat

Keyword(s):

Time Series ◽

Anomaly Detection ◽

Autoregressive Model ◽

Gas Turbines ◽

Time Series Data ◽

Combined Cycle ◽

Series Data ◽

Vector Autoregressive Model ◽

Vector Autoregressive ◽

Multi Stage

Abstract This study is based on time-series data from the combined cycle utility gas turbines consisting of three-gas turbine units and one steam turbine unit. We construct a multi-stage vector autoregressive model for the nominal operation of powerplant assuming sparsity in the association among variables and use this as a basis for anomaly detection and prediction. This prediction is compared with the time-series data of the plant-operation containing anomalies. Granger causality networks, which are based on the associations between the time series streams, are learned as an important implication from the vector autoregressive modelling. Anomaly is detected by comparing the observed measurements against their predicted value.

Download Full-text