OLS Estimates Research Articles

Panel cointegrating polynomial regression analysis and an illustration with the environmental kuznets curve

The analysis of cointegrating polynomial regressions, i.e, regressions that include an integrated process and its powers as explanatory variables is extended from the time series to the panel case by developing two estimators, a modified and a fully modified OLS estimator. As usual in the cointegration literature, the stationary errors are allowed to be serially correlated and the regressors are allowed to be endogenous. Both individual and time fixed effects are accommodated and the analysis uses an i.i.d. random linear process framework. The modified OLS estimator utilizes the large cross-sectional dimension that allows to consistently estimate and subtract an additive bias term without the need to also transform the dependent variable as required in fully modified OLS estimation. Both developed estimators have zero mean Gaussian limiting distributions and thus allow for standard asymptotic inference. A brief application to the environmental Kuznets curve illustrates the developed methods.

Methods for Modeling Autocorrelation and Handling Missing Data in Mediation Analysis in Single Case Experimental Designs (SCEDs).

Single-Case Experimental Designs (SCEDs) are increasingly recognized as a valuable alternative to group designs. Mediation analysis is useful in SCEDs contexts because it informs researchers about the underlying mechanism through which an intervention influences the outcome. However, methods for conducting mediation analysis in SCEDs have only recently been proposed. Furthermore, repeated measures of a target behavior present the challenges of autocorrelation and missing data. This paper aims to extend methods for estimating indirect effects in piecewise regression analysis in SCEDs by (1) evaluating three methods for modeling autocorrelation, namely, Newey-West (NW) estimation, feasible generalized least squares (FGLS) estimation, and explicit modeling of an autoregressive structure of order one (AR(1)) in the error terms and (2) evaluating multiple imputation in the presence of data that are missing completely at random. FGLS and AR(1) outperformed NW and OLS estimation in terms of efficiency, Type I error rates, and coverage, while OLS was superior to the methods in terms of power for larger samples. The performance of all methods is consistent across 0% and 20% missing data conditions. 50% missing data led to unsatisfactory power and biased estimates. In light of these findings, we provide recommendations for applied researchers.

Earnings penalty of educational mismatch: a comparison of alternative methods of assessing over-education

In this paper we systematically evaluate the impact of using the alternative methods conventionally used in the international literature on the measured incidence of educational mismatch and its earnings effects. We use a rich Australian longitudinal data set for a controlled group of full-time employed workers. Using panel data estimation, we address individual heterogeneity and measurement error, which are important in educational mismatch analysis. We show that alternative methods of measurement result in a range of estimates, with the Mode measure providing the most stable results across instrumental variables (IV) selections in panel fixed effects instrumental variables (FEIV) estimations. Based on the Mode measure, the incidence rate of over-education is 32.3%. The earnings penalty for each year of over-education is 2.5%, which is larger than 0.6% in fixed effect estimation and also larger than 1.9% in OLS estimations.

The Effect of Working Mother Status on Children’s Education Attainment: Evidence from Longitudinal Data

This study builds on the debate on whether mothers‘ employment in the early life of children bring positive or negative impact to children in later life. This is based on the fact that employment often result in reduced family time in which mother may not be able to take care of their children fully. This study estimated the effects of mothers’ employment status on children’s education in the short-term and in the long-term in 24 provinces in Indonesia. This study used data on children from three waves of the Indonesia Family Life Survey (IFLS): IFLS-3 (in 2000) aged 0–7 years, IFLS-4 (in 2007) aged 7–14 years, and IFLS-5 (in 2014) aged 14–21 years. The outcome variable was the children’s education and the variable of interest was the mother’s employment status. To overcome endogeneity, the analysis of the relationship between outcome and interest variables in this study was done by using the ordinary least square estimation (OLS) method and instrumental variables (IV). This study also used a cross-sectional design which estimated IFLS-4 (in 2007) and IFLS-5 (in 2014), separately. The main finding suggests that the mother’s employment status positively influenced children’s education both in the short- and long-term, as evidenced by the ordinary least square estimation (OLS) results. The instrumental variable (IV) estimate showed that decision-making was a strong instrument; and, it further revealed that mothers’ decision-making within the family affected their opportunity to work. This research could contribute to strengthening working mothers’ self-border and the concept of work-family facilitation in a family. It could also become a reference for stakeholders involved in the policy making to regulate policies which facilitate and support working mother to create ideal working environment. This condition is expected to supports children’s development as well as creating equity for working mothers in Indonesia.

Improved high-dimensional regression models with matrix approximations applied to the comparative case studies with support vector machines

Nowadays, high-dimensional data appear in many practical applications such as biosciences. In the regression analysis literature, the well-known ordinary least-squares estimation may be misleading when the full ranking of the design matrix is missed. As a popular issue, outliers may corrupt normal distribution of the residuals. Thus, since not being sensitive to the outlying data points, robust estimators are frequently applied in confrontation with the issue. Ill-conditioning in high-dimensional data is another common problem in modern regression analysis under which applying the least-squares estimator is hardly possible. So, it is necessary to deal with estimation methods to tackle these problems. As known, a successful approach for high-dimension cases is the penalized scheme with the aim of obtaining a subset of effective explanatory variables that predict the response as the best, while setting the other parameters to zero. Here, we develop several penalized mixed-integer nonlinear programming models to be used in high-dimension regression analysis. The given matrix approximations have simple structures, decreasing computational cost of the models. Moreover, the models are effectively solvable by metaheuristic algorithms. Numerical tests are made to shed light on performance of the proposed methods on simulated and real world high-dimensional data sets.

Empirical research methods department: Mission, learnings, and future plans

Empirical research methods department: Mission, learnings, and future plans

Does Organizational Culture Impact on Firm Performance: Evidence From Turkey

This paper aims to examine the impact of corporate culture over the financial performance of certain enterprises therewith by covering the data of chemical, petroleum, rubber and plastic products firms (XCHEMST) traded in Borsa Istanbul. In that regard, corporate culture data have been obtained from the surveys of the Denison’s Organizational Culture Model (DOCM) employees and managers of these businesses while the financial data were gleaned from financial statements of the mentioned enterprises between 2012-2016. We posited the ROE as a dependent variable while ROA and DEBT EQUITY have been employed as explanatory financial variables. Accordingly, we have determined SIZE and AGE as firm-specific variables. Utilizing cross-sectional data and multiple linear regression models, we estimated the coefficients of the variables with the aid of an OLS estimator. In so doing, this paper provides sufficient empirical evidence that three out of four cultural traits in the Denison’s Theory have had no impact on the financial performance of Turkey’s chemical firms traded on BIST, whereas the mission trait has had a significant impact. Consequently, our results are largely in line with the findings of other C/P research, suggesting that the mission trait of the DOCM is the core cultural term in ensuring thriving firm corporate performance.

Parental Discipline and Early Childhood Development in Rural China

Children who are under the age of five in underdeveloped and developing countries, including China, exhibit developmental delays due to their exposure to risks such as impoverishment, deprived health conditions, parental punishment, neglect, and poor psychosocial stimulation. Adverse experiences during the formative years of life, such as harsh parental discipline, may put them at risk for poor physical and mental well-being. The aim of this research is to explore the pervasiveness and developmental outcomes of different forms of discipline practices in the underdeveloped rural areas of China. To do this, we used cross-sectional data on child–caregiver dyads from a large survey held in 22 poor counties in the QinBa Mountain Region. The sample included 1622 children aged 12–36 months. Partakers were requested to respond to a general survey on parenting which included basic demographic questions, the Ages and Stages Questionnaire: Social-Emotional (ASQ-SE), the Bayley Scales of Infant and Toddler Development (BSID-III), and questions on exposure of children to different discipline practices. Our findings from OLS estimates reveal that aversive discipline methods adversely affect cognitive, language, and socio-emotional development, whereas, non-aversive discipline practices have constructive effects on cognitive, language, and motor development of children.

Average ordinary least squares‐centeredpenalized regression: A more efficient way to address multicollinearity than ridge regression

AbstractWe developed a novel method to address multicollinearity in linear models called average ordinary least squares (OLS)‐centered penalized regression (AOPR). AOPR penalizes the cost function to shrink the estimators toward the weighted‐average OLS estimator. The commonly used ridge regression (RR) shrinks the estimators toward zero, that is, employs penalization prior in the Bayesian view, which contradicts the common real prior . Therefore, RR selects small penalization coefficients to relieve such a contradiction and thus makes the penalizations inadequate. Mathematical derivations remind us that AOPR could increase the performance of RR and OLS regression. A simulation study shows that AOPR obtains more accurate estimators than OLS regression in most situations and more accurate estimators than RR when the signs of the true s are identical and is slightly less accurate than RR when the signs of the true s are different. Additionally, a case study shows that AOPR obtains more stable estimators and stronger statistical power and predictive ability than RR and OLS regression. Through these results, we recommend using AOPR to address multicollinearity more efficiently than RR and OLS regression, especially when the true s have identical signs.

Does Fuel Pricing Affect Males and Females Differently? Evidence from Kerosene Using Households in Rural Bangladesh

Energy sector reform, particularly the pricing of fuel and its gender-differentiated impacts, is rarely studied in the literature on energy and gender. This empirical study examines whether energy pricing has differential consequences for men and women with regard to kerosene oil use in rural Bangladesh. Specifically, the variables under focus are the duration of the study, duration of night-time leisure and time spent on income-generating activities. A household survey with 630 kerosene using rural households has been conducted in eight districts of Bangladesh. Apart from OLS estimation, we also use IV to address the endogeneity embedded in the model. The study finds that higher kerosene prices have a gender-differentiated effect on study duration, night-time leisure duration and time spent on income-generating activities in the context of household use of lighting fuel. Females are more adversely affected by higher kerosene prices in terms of study duration and night-time leisure hours. The effort to offset these losses by increasing effort on income-generating activities is less successful for women than men.

WITHDRAWN: Monetary policy channels, sectoral outputs and sustainable growth in ECOWAS region: A rigorous analysis and implications for policy

The market-based monetary policy framework has been favoured by ECOWAS economies. Hence this study investigated the effect of monetary policy channels on the sectoral value added and sustainable economic growth in ECOWAS. Data from World Bank and International Monetary Fund over 2013 – 2019 were sourced for thirteen member countries. ECOWAS is found to have very high inflation level, interest and exchange rates. Furthermore, using the Driscoll-Kraay fixed-effects OLS estimator, the findings revealed that while the effect of monetary policy channels on the agricultural sector value added is largely heterogenous and significantly inelastic, the one on the industrial and services sectors are overwhelmingly homogeneous and negative, but insignificant for the services sector. Moreover, the effect of monetary policy channels on sustainable economic growth is also homogeneously asymmetric, with imminent stagflation, while the interactive effects of monetary policy channels are heterogeneous on sustainable economic growth and economic sectors. Therefore, an inflation targeting monetary policy stance is generally recommended with prioritised exchange rate stabilisation amid sufficient fiscal space.

Challenges for sustainable productivity in developing economies: shortage of energy and corruption

ABSTRACT Developing countries suffer from shortages of electricity, posing a severe problem for industrial firms and giving them considerable incentives to access electricity resources through bribery. We investigate how firms’ revenues, sales and productivity are affected by payments of bribes to obtain electricity connections. We use ordinary least squares (OLS), quantile and instrumental variable (IV) quantile regression analysis. OLS estimates reveal a negative effect on sales and productivity and a positive impact on profitability, while quantile regression suggests a negative effect on sales across all quantiles and a positive effect on productivity at the lower quantile. Since this research shows that bribery may negatively influence productivity, it informs shareholders that bribery may be detrimental to their long-term benefits.

Land Inequality and Workfare Policies

This paper contributes to the relatively scant literature on the impacts of inequality on the efficacy of public works programmes. We study this in the context of India. In particular, we examine the effect of land inequality on the implementation of the world’s largest workfare programme – the National Rural Employment Guarantee Act (NREGA). Our OLS estimates demonstrate that the concentration of land ownership reduces the efficacy of NREGA. An instrumental variable (IV) analysis, where we use the historical land tenure system as an IV for contemporaneous land inequality, further corroborates our findings. This negative relationship is consistent with the hypothesis that public work schemes raise agricultural wages in the private labour market, thereby incentivising big landlords to use their political power to oppose such programmes. We exclude the possibility that the higher provision of public jobs in more equal areas is driven by a higher demand for public jobs or by caste or religious differences. This study suggests that the concentration of land ownership, a proxy for power asymmetries, could hinder effective implementation of development policies.

Air pollution and tourism growth relationship: exploring regional dynamics in five European countries through an EKC model.

The present study intends to explore the relationship between tourism growth and air pollution at a regional level for five important tourism European destinations: France, Spain, Greece, Portugal, and Italy. Most of the studies found in the literature examine this relationship on a national scale and focus only on the CO2 pollutant, which is a greenhouse gas but not a critical pollutant in terms of air quality and human exposure. This research focuses on a regional basis (NUTS 2 classification) and takes into account the main critical pollutants in terms of urban air pollution (namely: NOx, PM10, and PM2.5), and considers 10 years, from 2009 until 2018. This work aims to investigate evidence of a tourism-induced Environmental Kuznets Curve (EKC) for the countries through the construction of five panels, one for each country, including different variables: the Gross Domestic Product, the energy consumption, and the number of nights spent at tourist accommodation establishments from both domestic and foreign tourists. The Levin-Lin-Chu unit root test proves the variables to be stationary, while the Pedroni cointegration test shows that they are integrated. The pooled OLS estimator is employed throughout the countries to check the relationship among the variables. Results reveal that the tourism-induced EKC hypothesis is not validated for any of the countries. The findings also show that in Portugal, Italy, and Greece, there is a negative relationship between economic growth and environmental pollution, while mixed evidence is found for France and Spain. Moreover, differences in the impacts of international and domestic tourists on air pollution are found: foreign tourists negatively impact emissions, while domestic ones increase them. This result is clear for Spain, Greece, and Italy. The Granger panel causality test is then conducted to see the causality among the variables.

Should we care (more) about data aggregation?

Aggregation tools transform multidimensional data into indices. To investigate how the design of an aggregation process affects regression results, we build democracy indices that differ regarding their scale and aggregation function. Using the democracy–growth nexus as a testing ground, we illustrate that the choice of the aggregation procedure significantly affects OLS and 2SLS estimates since different methods produce systematically different index values for observations at the lower and upper end of the autocracy–democracy spectrum. We also illustrate that dichotomous measures produce significantly smaller OLS estimates than continuous measures due to lower discriminating power. Whether continuous and dichotomous indicators create different 2SLS estimates depends on their design. Because of the methodological similarities of democracy indicators and other social science indicators, we expect similar consequences for other empirical analyses.

Systems estimation of a structural model of distribution and demand in the US economy

Abstract This paper investigates non-cyclical, short-run relationships between income distribution and the components of aggregate demand in the US from 1963–2016. Previous studies using this ‘structural’ methodology have typically found that demand is wage-led in most large, advanced economies. However, these studies have been criticised for treating total output and the wage share as exogenous, potentially leading to simultaneity bias. This paper corrects for such possible bias as well as common shocks to the equations by using systems GMM. Surprisingly, these estimates imply that private-sector aggregate demand is more, rather than less, wage-led (or in some cases, less profit-led) compared with OLS estimates of identically specified models. This paper is also the first to provide separate estimates of non-residential and residential investment functions and to distinguish the effects of shocks to different underlying determinants of the wage share (unit labour costs and firms’ monopoly power), finding that these differ qualitatively.

A Silent Crisis: The Impact of Public Health Expenditure on Malnutrition Prevalence in Children Aged Below Five in the Philippines

This study mainly aims to determine whether public health expenditures have been effective in reducing malnutrition among children aged below five in the Philippines. The researchers construct a Grossman (1972) model-based health production function, which treats economic, social, and environmental factors as determinants of nutritional status. OLS estimates show that an increase in food security rates, a decrease in poverty incidence rates, and an increase in the level of urbanization significantly reduce stunting rates. However, no statistically significant relationship exists between the aforementioned independent variables and underweight and wasting rates (aside from the level of urbanization and wasting). In all regression models, the coefficient estimate for public health expenditure is valued near zero and is statistically insignificant, implying that government spending on health has been insubstantial and ineffective in reducing malnutrition prevalence.

Trend Analysis and Prediction on Water Consumption in Southwestern Ethiopia

This study intensively examined the monthly water consumption forecasting performance using advanced time series (ARIMA) models. Thus, this study intends to identify the appropriate ARIMA models to best fit the water consumption data in Southwestern Ethiopia Tepi town and forecast water consumption effectively in the city. The data used for this study was the monthly water consumption in Tepi town from January 2016 to December 2021.The data were converted to returns to enhance their statistical properties and the returns were used to fit a mean equation. The monthly average water consumption in Tepi Town is 77227.8 meters cubic. Both original and transformed data show the trend of water consumption is increasing over time. Several ARIMA models were fitted to the data, and it emerged that the most adequate model for the data was ARIMA (1, 1, 1) based on the model selection criterion. The parameters for ARIMA models were estimated using the Ordinary Least Squares Estimation (OLS) method. The model was used to forecast the consumption for the next ten months and to advise Tepi town Water Company Limited in the city to meet the demand of the people. Conclusion. The consumption of water is increasing from December to September.

Robust Mediation Analysis: The R Package robmed

Mediation analysis is one of the most widely used statistical techniques in the social, behavioral, and medical sciences. Mediation models allow to study how an independent variable affects a dependent variable indirectly through one or more intervening variables, which are called mediators. The analysis is often carried out via a series of linear regressions, in which case the indirect effects can be computed as products of coefficients from those regressions. Statistical significance of the indirect effects is typically assessed via a bootstrap test based on ordinary least-squares estimates. However, this test is sensitive to outliers or other deviations from normality assumptions, which poses a serious threat to empirical testing of theory about mediation mechanisms. The R package robmed implements a robust procedure for mediation analysis based on the fast-and-robust bootstrap methodology for robust regression estimators, which yields reliable results even when the data deviate from the usual normality assumptions. Various other procedures for mediation analysis are included in package robmed as well. Moreover, robmed introduces a new formula interface that allows to specify mediation models with a single formula, and provides various plots for diagnostics or visual representation of the results.

Finite Sample Identification of Low-Order LTI Systems via Nuclear Norm Regularization

This paper studies the problem of identifying low-order linear time-invariant systems via Hankel nuclear norm (HNN) regularization. This regularization encourages the Hankel matrix to be low-rank, which corresponds to the dynamical system being of low order. We provide novel statistical analysis for this regularization, and contrast it with the unregularized ordinary least-squares (OLS) estimator. Our analysis leads to new finite-sample error bounds on estimating the impulse response and the Hankel matrix associated with the linear system using HNN regularization. We design a suitable input excitation, and show that we can recover the system using a number of observations that scales optimally with the true system order and achieves strong statistical estimation rates. Complementing these, we also demonstrate that the input design indeed matters by proving that intuitive choices, such as i.i.d. Gaussian input, lead to sub-optimal sample complexity. To better understand the benefits of regularization, we also revisit the OLS estimator. Besides refining existing bounds, we experimentally identify when HNN regularization improves over OLS: (1) For loworder systems with slow impulse-response decay, OLS method performs poorly in terms of sample complexity, (2) the Hankel matrix returned by regularization has a more clear singular value gap that makes determining the system order easier, (3) HNN regularization is less sensitive to hyperparameter choice. To choose the regularization parameter, we also outline a simple joint train-validation procedure