Reichenbach’s ‘principle of the common cause’ is a foundational assumption of some important recent contributions to quantitative social science methodology but no similar principle appears in econometrics. Angrist et al. has argued that the principle is necessary for instrumental variables methods in econometrics, and Angrist Krueger builds a framework using it that he proposes as a means of resolving an important methodological dispute among econometricians. Through analysis of instrumental variables methods and the issue of multicollinearity, we aim to show (...) that the relationship of the principle to econometric methods is more nuanced than implied by previous work but nevertheless may make a valuable contribution to the coherence and validity of existing methods. (shrink)

The evidence on gender discrimination in lending remains controversial. To capture gender biases in banks’ loan allocations, we observe the impact on the applicants of a microfinance institution and exploit the natural experiment of a regulatory change imposing a strict EUR 10,000 loan ceiling on microcredit. Descriptive statistics indicate that the presence of the ceiling is associated both with bank-MFI co-financing and with harsher treatment of female borrowers. To investigate causal links, we develop an econometric approach that addresses the concerns (...) of selection biases, multicollinearity, and endogeneity. Our empirical findings suggest that the change in the MFI’s gender-related attitude was triggered by banks through co-financing. Hence, we speculate that co-financing pushes ceiling-constrained MFIs to import whatever biases in loan granting that the banks are prone to. Overall, this paper stresses that apparently benign regulations such as loan ceilings can significantly harm the women’s empowerment efforts made by MFIs. (shrink)

This work compares Autometrics with dual penalization techniques such as minimax concave penalty and smoothly clipped absolute deviation under asymmetric error distributions such as exponential, gamma, and Frechet with varying sample sizes as well as predictors. Comprehensive simulations, based on a wide variety of scenarios, reveal that the methods considered show improved performance for increased sample size. In the case of low multicollinearity, these methods show good performance in terms of potency, but in gauge, shrinkage methods collapse, and higher (...) gauge leads to overspecification of the models. High levels of multicollinearity adversely affect the performance of Autometrics. In contrast, shrinkage methods are robust in presence of high multicollinearity in terms of potency, but they tend to select a massive set of irrelevant variables. Moreover, we find that expanding the data mitigates the adverse impact of high multicollinearity on Autometrics rapidly and gradually corrects the gauge of shrinkage methods. For empirical application, we take the gold prices data spanning from 1981 to 2020. While comparing the forecasting performance of all selected methods, we divide the data into two parts: data over 1981–2010 are taken as training data, and those over 2011–2020 are used as testing data. All methods are trained for the training data and then are assessed for performance through the testing data. Based on a root-mean-square error and mean absolute error, Autometrics remain the best in capturing the gold prices trend and producing better forecasts than MCP and SCAD. (shrink)

Introduction: Motor evoked potentials (MEPs) to transcranial magnetic stimulation are known to be susceptible to several sources of variability. However, conflicting evidences on individual characteristics in relatively small sample sizes have been reported. We investigated the effect of age, height, and sex on MEPs of the motor cortex and spinal roots in a large cohort. Methods: A total of 587 subjects clinically and neuroradiologically intact were included. MEPs were recorded during mild tonic contraction through a circular coil applied over the (...) “hot spot” of the first dorsal interosseous and tibialis anterior muscles, bilaterally. Central motor conduction time (CMCT) was estimated as the difference between MEP cortical latency and the peripheral motor conduction time by cervical or lumbar magnetic stimulation. Peak-to-peak MEP amplitude to cortical stimulation and right-to-left difference of each parameter were also measured. Results: After Bonferroni correction, general linear (multiple) regression analysis showed that both MEP cortical latency and peripheral motor conduction time at four limbs positively correlated with age and height. At lower limbs, an independent effect of sex on the same measures was also observed (with females showing smaller values than males). CMCT correlated with both age (negatively) and height (positively) when analyzed by a single regression; however, with a multiple regression analysis this significance disappeared, due to the correction for the multicollinearity within the dataset. Conclusions: Physical individual features need to be considered for a more accurate and meaningful MEPs interpretation. Both in clinical practice and in research setting, patients and controls should be matched for age, height, and sex. (shrink)

In terms of scenic marketing, big data research also plays an important role in the precise marketing of scenic spots. This paper has focused on the big data related to scenic spots as the research object, explores the relationship between various subdivision big data and the number of tourists in scenic spots, and investigates the difference and influence of the consumption behavior of the secondary consumption items in the scenic area, to find the potential of the scenic area’s business growth (...) and to promote the continuous and stable growth of the scenic area’s sales and tourism economy. Using the relevant theories and analysis methods, such as consumer behavior, big data, and tourism consumer behavior, the content mainly focuses on the establishment of the analysis model of the number of tourists in the scenic spot, the data collection, the estimation of the model parameters, the various types of big data, the calculation of the contribution rate of the data to the number of tourists in the scenic spot, and the difference analysis of the secondary consumption items of different types of tourists in the scenic spot. Results show that a multi-objective analysis model is established based on the relevant econometric theories, and an optimization plan is proposed after the multicollinearity diagnosis of the model; to establish a data envelopment analysis model of the difference and influence of different types of tourists’ consumption behavior in scenic spots and study the consumption behavior characteristics of different types of tourists when they purchase secondary consumption items in scenic spots; the econometric model is used to analyze the big data, adjust the linear relationship of some variables, then adopt the method of gradually adding variables combined with the consumer theory, and finally determine the number of daily tourists as the explained variable, the number of internet protocol, Baidu index, and the virtual value of the weekend, dummy variables for variables, bounce rate, and air pollution as explanatory variables. (shrink)

Significant research has been devoted to detecting people in images and videos. In this paper we describe a human de- tection method that augments widely used edge-based fea- tures with texture and color information, providing us with a much richer descriptor set. This augmentation results in an extremely high-dimensional feature space (more than 170,000 dimensions). In such high-dimensional spaces, classical machine learning algorithms such as SVMs are nearly intractable with respect to training. Furthermore, the number of training samples is much (...) smaller than the dimensionality of the feature space, by at least an order of magnitude. Finally, the extraction of features from a densely sampled grid structure leads to a high degree of multicollinearity. To circumvent these data characteristics, we employ Partial Least Squares (PLS) analysis, an effi- cient dimensionality reduction technique, one which pre- serves significant discriminative information, to project the data onto a much lower dimensional subspace (20 dimen- sions, reduced from the original 170,000). Our human de- tection system, employing PLS analysis over the enriched descriptor set, is shown to outperform state-of-the-art tech- niques on three varied datasets including the popular INRIA pedestrian dataset, the low-resolution gray-scale Daim- lerChrysler pedestrian dataset, and the ETHZ pedestrian dataset consisting of full-length videos of crowded scenes. (shrink)

The unprecedented economic growth in recent decades has cultivated the exploitation of natural resources and over-consumption, leading to ecological deterioration and sustainability. The ever-increasing consumption in developing countries is creating a significant environmental strain. Thus, the industry and consumers’ environmental issues and their harmful effects on human health have led to concerns among researchers, scientists, academic communities, and policymakers. The present work examines the impact of different consumption value factors on sustainable consumption behavior concerning consumer choice in Pakistan and China. (...) A cross-sectional study is conducted, and data are collected through a primary source questionnaire. A sample of 431 respondents is chosen from different cities in Pakistan, and a sample of 342 respondents is selected from China. Estimation techniques like descriptive statistics, frequency distribution, multicollinearity, R square, independent sample t-test, the coefficient of correlation, and regression analysis are used for the data analysis. The comparative results show that knowledge values and emotional values significantly influence the choice behavior of respondents toward environmentally friendly products both in Pakistan and China. In contrast, social values and conditional values show insignificant influence. Furthermore, functional values are significant in Pakistan while insignificant in the context of China, and environmental values are significant in China although insignificant in Pakistan with regard to sustainable consumption behavior. (shrink)

This paper uses the Multidimensional Fatigue Inventory to investigate the fatigue status of pilots, and the reliability and validity of the scale are tested by Cronbach’s α and exploratory factor analysis. The founding shows that mild fatigue and above accounted for 67.7%. For further quantify the impact of different flights on pilots’ fatigue, research improves the fatigue coefficient model based on the results of pilot fatigue feeling questionnaire. Combined with multifactor analysis of variance and multiple linear regression, it is found (...) that the independent variables have different and positive effects on the dependent variables, and there is no multicollinearity. Through the actual test, its accuracy is improved by 16.7% compared with the original model. (shrink)

Dual innovation, which includes exploratory innovation and exploitative innovation, is crucial for firms to obtain a sustainable competitive advantage. The knowledge base of firms greatly influences or even determines the scope, direction, and path of their dual-innovation activities, which drive their innovation process and produce different innovation performances. This study uses data source patents obtained by 285 focal firms in the Chinese new-energy vehicle industry in the period 2015–2020. Five knowledge-base features are selected by analyzing the correlation and multicollinearity, (...) and four different firm clusters are found by using the k-means clustering algorithm. Based on the classification and regression tree algorithm, we mine the potential decision rules governing the dual-innovation performance of firms. The results show that the exploratory innovation performance of firms in different clusters is mainly affected by two different knowledge-base features. Knowledge-base scale is a key factor affecting the exploitative innovation performance of firms. Firms in different clusters can improve their dual-innovation performance by rationally tuning the combination of knowledge-base features. (shrink)