China 
Africa 
Explore chapters and articles related to this topic
Recognizing risk factors associated with crash frequency on rural four lane highways
Published in Sandra Erkens, Xueyan Liu, Kumar Anupam, Yiqiu Tan, Functional Pavement Design, 2016
C. Naveen Kumar, M. Parida, S.S. Jain
The Mean Squared Prediction Error (MSPE) is a traditional indicator of error and calculates the difference between the estimated and observed values squared.() MSPE=1n∑i=1n(μi^−yi)2
A closed-form moment estimator for the vector multiplicative error model and its application
Published in Quality Technology & Quantitative Management, 2019
Wanbo Lu, Yanfeng Wang, Jungong Li, Rui Ke
We will consider the CLFE and the QLME of the vMEM(1,1) of the absolute return and the high-low range in this subsection. We perform the Augmented Dickey Fuller (ADF) test for the absolute return and the high–low range. These two series are both stationary since the statistics of the ADF are −7.2277 and −6.7794, respectively, which are smaller than the critical values at 1, 5 and 10% significant level. We also find that the coefficient of autocorrelation and partial autocorrelation decreases with increasing lag order. The Ljung-Box test statistics increases with increasing lag order. Thus, we estimate the parametric vMEM(1,1) for these two series. To evaluate the fitting and forecasting accuracy, mean squared prediction error (MSPE) is used in the in-sample period and the out-of-sample period. The assessment index is defined as
Robust Lasso Regression Using Tukey's Biweight Criterion
Published in Technometrics, 2018
Le Chang, Steven Roberts, Alan Welsh
Variable selection accuracy is measured by the number of correctly identified significant variables (No. correct), the number of included noise variables (No. incorrect) and the percentage of correctly fitted models (correctly fitted). The percentage of correctly fitted models is the proportion of times that the selected model includes all significant variables and excludes all noise variables over 100 simulations. Prediction accuracy is measured by the mean squared prediction error (MSPE) , computed over a set of independent test samples with the same sample size n as the training sample. We conduct 100 repeated simulations and compute the average, the median and the standard error of the MSPEs.