Zapranis, A. and Refenes, A.P.N. (1999) Model Adequacy Testing. In: Principles of Neural Model Identification, Selection and Adequacy. In: Perspectives in Neural Computing, Springer, London.
https://doi.org/10.1007/978-1-4471-0559-6

has been cited by the following article:

• TITLE: White Noise Analysis: A Measure of Time Series Model Adequacy

  AUTHORS: Imoh Udo Moffat, Emmanuel Alphonsus Akpan

  KEYWORDS: Autocorrelation, Model Identification, Model Estimation, Diagnostic Checking, Time Series

  JOURNAL NAME: Applied Mathematics, Vol.10 No.11, November 21, 2019

  ABSTRACT: The purpose of this study is to apply white noise process in measuring model adequacy targeted at confirming the assumption of independence. This ensures that no autocorrelation exists in any time series under consideration, and that the autoregressive integrated moving average (ARIMA) model entertained is able to capture the linear structure in such series. The study explored the share price series of Union bank of Nigeria, Unity bank, and Wema bank obtained from Nigerian Stock Exchange from January 3, 2006 to November 24, 2016 comprising 2690 observations. ARIMA models were used to model the linear dependence in the data while autocorrelation function (ACF), partial autocorrelation function (PACF), and Ljung-Box test were applied in checking the adequacy of the selected models. The findings revealed that ARIMA(1,1,0) model adequately captured the linear dependence in the return series of both Union and Unity banks while ARIMA(2,1,0) model was sufficient for that of Wema bank. Also, evidence from ACF, PACF and Ljung-Box test revealed that the residual series of the fitted models were white noise, thus satisfying the conditions for stationarity.