China
Africa
Explore chapters and articles related to this topic
Identifying Non-linearity in Construction Workers' Personality Safety Behaviour Predictive Relationship Using Neural Network and Linear Regression Modelling
Published in M.Z. Naser, Leveraging Artificial Intelligence in Engineering, Management, and Safety of Infrastructure, 2023
Yifan Gao, Vicente A. González, Wing Tak Yiu, Guillermo Cabrera-Guerrero
In addition, it has been pointed out that an LR generally has a relatively simple mathematical structure compared with nonlinear techniques such as NN (Velasco et al., 2020). Such a simple structure may lead to instability of the regression equations, being sensitive to potential structural breaks in the input signals (Peters et al., 2019). A structural break refers to an abrupt shift in the slope of a trend line over a series of data points. To ascertain the structural stability of the LR formulas (2.5) and (2.6), a widely used approach—Chow test—was adopted. The Chow test can estimate whether the parameters of an LR model are structurally stable, and runs as follows (Song et al., 2019): 1. Identifying the structural breaks in the dataset used to develop the model and splitting the dataset into subsets at the breakpoints; 2. Performing separate regressions on the entire dataset and each subset of the data; 3. Retrieving the residual sum of squares for each regression; and 4. Computing the Chow statistic using the formula: F=(N−2k)[Rw−(∑j=1nRj)]k(∑j=1nRj) where F = the Chow statistic; Rw= residual sum of squares of the regression for the whole dataset; n = the number of sub-datasets split according to structural breaks; j = the jth sub-dataset; Rj= residual sum of squares of the regression for the jth sub-dataset; N = the number of samples in the whole dataset (N = 228, including 188 training samples and 40 validating samples as assigned in Section The Data Split Ratio); and k = the number of input variables (k = 4, including four personality traits).
Forecasting the determinants of environmental degradation: a gray modeling approach
Published in Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, 2020
Aliya Shaheen, Jinyong Sheng, Sadia Arshad, Hafeez Muhammad, Shafaq Salam
The study will not rely only on conventional unit-root techniques. The traditional unit-root tests do not offer information concerning structural breaks in the data series (Shahbaz et al. 2014). Therefore, we will also perform unit-root test with structural breaks. The existence of structural breaks cannot be ignored in the presence of an extended data series. We applied Zivot and Andrew structural break unit-root test to consider structural variations in the data (Zivot and Andrews 2002). The outcomes of the test are stated in Table 5. The two variables N2O emissions and agriculture are stationary at level with the structural break in 1982 and 1983, respectively. The CO2 emissions, energy consumption, GDP, and urbanization are stationary at first difference with structural breaks in 2008, 2007, 1992, and 1980 correspondingly. The reasons for breaks in data may be due to various causes, i.e. political factors, natural calamities, growth in industrial production, increase in usage of fertilizers, better employment, and other services in urban centers, and global economic circumstances. Hence, the null hypothesis of the unit root is rejected.
On the persistence of sectoral electricity consumption in Iran: a GARCH-based unit root approach
Published in Energy Sources, Part B: Economics, Planning, and Policy, 2020
Iman Cheratian, Saleh Goltabar, Antonio A. Golpe
where . Next, it is required to identify two structural breaks. The unknown break dates are determined using the Bai and Perron (2003) multiple structural break test. The favorable feature of the Bai and Perron (2003) test is its ability to search for a maximum of five structural breaks in a time series.2For more information, see Narayan and Liu (2015). The unknown break dates are measured by a sequential application of the sup FT (+1|) test. The first structural break =1 with the break date TB1 is estimated when the null hypothesis for F-statistic sup FT (+1|) is rejected. Therefore, the first break date is estimated using the maximum absolute t-value of the break dummy coefficient D1, as follows:
Electricity consumption and economic growth nexus in Zimbabwe revisited: fresh evidence from Maki cointegration
Published in International Journal of Green Energy, 2019
Remember Samu, Festus Victor Bekun, Murat Fahrioglu
The preliminary visual plot of series is necessary for the time series econometric analysis. This is in order to have a glimpse of the nature of variables. Figure 1 presents the graphical plot in their natural logarithm form. The graphical plot shows the possibility of a structural break(s). Thus, it is pertinent to account for such breaks in the econometric analysis to avoid spurious analysis. To this end, the current study utilizes estimation techniques that account for a structural break(s). Table 2 reports the summary statistics and correlation matrix for the series under review. All series are normally distributed except for electricity consumption. Also seen is the negative skewness exhibited by all series. The results of the correlation matrix are very insightful. There exists a positive and significant relationship between economic growth and carbon dioxide emissions. This is true for the case study. Also revealed by the correlation matrix is the significant positive association between electricity consumption and economic growth. This is obvious for Zimbabwe given her swift growing population. Although correlation analysis gives a glimpse of the sort of relationship between variables. However, the correlation analysis is not sufficient to substantiate our claims. It is of paramount importance that this study proceeds to conduct more econometrics analysis to either refute or validate our study arguments.