Spatial-Temporal 3D Residual Correlation Network for Urban Traffic Status Prediction

Accurate traffic status prediction is of great importance to improve the security and reliability of the intelligent transportation system. However, urban traffic status prediction is a very challenging task due to the tight symmetry among the Human–Vehicle–Environment (HVE). The recently proposed spatial–temporal 3D convolutional neural network (ST-3DNet) effectively extracts both spatial and temporal characteristics in HVE, but ignores the essential long-term temporal characteristics and the symmetry of historical data. Therefore, a novel spatial–temporal 3D residual correlation network (ST-3DRCN) is proposed for urban traffic status prediction in this paper. The ST-3DRCN firstly introduces the Pearson correlation coefficient method to extract a high correlation between traffic data. Then, a dynamic spatial feature extraction component is constructed by using 3D convolution combined with residual units to capture dynamic spatial features. After that, based on the idea of long short-term memory (LSTM), a novel architectural unit is proposed to extract dynamic temporal features. Finally, the spatial and temporal features are fused to obtain the final prediction results. Experiments have been performed using two datasets from Chengdu, China (TaxiCD) and California, USA (PEMS-BAY). Taking the root mean square error (RMSE) as the evaluation index, the prediction accuracy of ST-3DRCN on TaxiCD dataset is 21.4%, 21.3%, 11.7%, 10.8%, 4.7%, 3.6% and 2.3% higher than LSTM, convolutional neural network (CNN), 3D-CNN, spatial–temporal residual network (ST-ResNet), spatial–temporal graph convolutional network (ST-GCN), dynamic global-local spatial–temporal network (DGLSTNet), and ST-3DNet, respectively.

Disaggregating the Effects of Inequality on Informal Giving: Evidence From Pakistan

In this article, we outline the determinants of informal charitable giving and the link between giving and inequality. Arguing that inequality encompasses at least two competing effects—distrust and observed need for donations—we use a novel proxy to separate out the effect of the latter from the former on household’s magnitude of informal giving. Using data from the Pakistan Centre for Philanthropy’s 2014 Indigenous Individual Philanthropy Survey, we find that informal giving in Pakistan follows patterns like those observed in the literature for formal giving. We also find evidence for a positive relationship between observed need and the magnitude of person-to-person giving. Controlling for observed need, we find that the residual correlation between inequality and giving is negative, one explanation of which may be the positive link between inequality and decreased social cohesion and trust.

COVID-19 mortality analysis from soft-data multivariate curve regression and machine learning

A multiple objective space-time forecasting approach is presented involving cyclical curve log-regression, and multivariate time series spatial residual correlation analysis. Specifically, the mean quadratic loss function is minimized in the framework of trigonometric regression. While, in our subsequent spatial residual correlation analysis, maximization of the likelihood allows us to compute the posterior mode in a Bayesian multivariate time series soft-data framework. The presented approach is applied to the analysis of COVID-19 mortality in the first wave affecting the Spanish Communities, since March, 8, 2020 until May, 13, 2020. An empirical comparative study with Machine Learning (ML) regression, based on random k-fold cross-validation, and bootstrapping confidence interval and probability density estimation, is carried out. This empirical analysis also investigates the performance of ML regression models in a hard- and soft-data frameworks. The results could be extrapolated to other counts, countries, and posterior COVID-19 waves.

The structural validity of the Finnish version of the Disabilities of the Arm, Shoulder and Hand: A Rasch model analysis

Introduction The construct validity of the Disabilities of the Arm, Shoulder and Hand questionnaire (DASH) has previously been questioned. The purpose of this study was to evaluate the measurement properties of the Finnish version of the DASH for assessing disability in patients with hand complaints using Rasch Measurement Theory. Methods A cohort of 193 patients with typical hand and wrist complaints were recruited at a surgery outpatient clinic. The DASH scores were analysed using the Rasch model for differential item functioning, unidimensionality, fit statistics, item residual correlation, coverage/targeting and reliability. Results In the original DASH questionnaire, the item response thresholds were disordered for 2 of 30 of the items. The item fit was poor for 9 of 30 of the items. Unidimensionality was not supported. There was substantial residual correlation between 87 pairs of items. Item reduction (chi square 95, degrees of freedom 50, p < 0.001) and constructing two testlets led to unidimensionality (chi square 0.64, degrees of freedom 4, p = 0.96). Person separation index was 0.95. The testlets had good fit with no differential item functioning towards age or gender. Conclusion Unidimensionality of the original Finnish version of the DASH was not supported, meaning the questionnaire seems to gauge traits other than disability alone. Hence, the clinician must be careful when trying to measure change in patients’ scores. Item reduction or the creation of testlets did not lead to good alternatives for the original Finnish DASH. Differential item functioning showed that the original Finnish scale exhibits minor response bias by age in one item. The original Finnish DASH covers different levels of ability well among typical hand surgery patients.

Validation of technical and vocational teachers’ competency evaluation instrument using the rasch model

This pilot study was conducted with the aim to validate the instrument used in evaluating the competency of teachers in the field of technical and vocational education. This instrument consists of 45 items and is administered to 53 teachers from a selected vocational college. The Rasch Model with the help of the Winstep Version 3.72.3 software has been used in this study for the purpose of checking the functionality of the item and the validity of the instrument. An analysis has been made based on the suitability of items in measuring the construct, item and person reliability and separation index, polarity and residual correlation value. The Rasch analysis showed that the item reliability was valued at 0.92 while the person reliability valued at 0.96 with their item MNSQ between overfit (<0.6) and misfit (>1.4). Based on the findings, there are three items that were dropped because of failing to meet the inspection criteria. The finalized instrument consists of 42 items, in which it is suitable for evaluating the four constructs in the competency evauation of technical and vocational teachers in vocational colleges.

PEMODELAN INDEKS HARGA KONSUMEN DI JAWA TENGAH DENGAN METODE GENERALIZED SPACE TIME AUTOREGRESSIVE SEEMINGLY UNRELATED REGRESSION (GSTAR-SUR)

The Generalized Space Time Autoregressive (GSTAR) model with Seemingly Unrelated Regression (SUR) estimation method or often called GSTAR-SUR is more efficient to be used for residual correlation than Ordinary Least Square (OLS) estimation method. The SUR estimation method utilizes residual correlation information to improve the estimated efficiency resulting in a smaller standard error. The purpose of this research is to get the GSTAR-SUR model according to Consumer Price Index (CPI) data in four regencies or cities in Central Java namely Purwokerto, Surakarta, Semarang, and Tegal. Based on the assumed white noise assumption, the smallest MAPE and RMSE averages, the best model chosen in this research is the GSTAR-SUR(11)I(1) model with the heavy of normalized cross-correlation with the average MAPE value of 0.4455% and RMSE value of 0.80582. The best model obtained explains that the CPI data in Purwokerto, Semarang, and Tegal not only influenced by the previous time but also influenced by the locations. Meanwhile, the CPI data in Surakarta is only influenced by the previous time, but it is not affected by other locations. Keywords: SUR, OLS, Consumer Price Index