THE IMPACT OF A HAUSMAN PRETEST ON THE ASYMPTOTIC SIZE OF A HYPOTHESIS TEST

2009 ◽  
Vol 26 (2) ◽  
pp. 369-382 ◽  
Author(s):  
Patrik Guggenberger
Keyword(s):  
Least Squares ◽  
Hypothesis Test ◽  
Ordinary Least Squares ◽  
Reduced Form ◽  
Structural Parameter ◽  
Ratio Test ◽  
Two Stage ◽  
Weak Instruments ◽  
Asymptotic Size ◽  
The Impact

This paper investigates the asymptotic size properties of a two-stage test in the linear instrumental variables model when in the first stage a Hausman (1978) specification test is used as a pretest of exogeneity of a regressor. In the second stage, a simple hypothesis about a component of the structural parameter vector is tested, using a t-statistic that is based on either the ordinary least squares (OLS) or the two-stage least squares estimator (2SLS), depending on the outcome of the Hausman pretest. The asymptotic size of the two-stage test is derived in a model where weak instruments are ruled out by imposing a positive lower bound on the strength of the instruments. The asymptotic size equals 1 for empirically relevant choices of the parameter space. The size distortion is caused by a discontinuity of the asymptotic distribution of the test statistic in the correlation parameter between the structural and reduced form error terms. The Hausman pretest does not have sufficient power against correlations that are local to zero while the OLS-based t-statistic takes on large values for such nonzero correlations. Instead of using the two-stage procedure, the recommendation then is to use a t-statistic based on the 2SLS estimator or, if weak instruments are a concern, the conditional likelihood ratio test by Moreira (2003).

scholarly journals Stein-like estimators for causal mediation analysis in randomized trials

2019 ◽  
Vol 29 (4) ◽  
pp. 1129-1148 ◽  
Author(s):  
Cedric E Ginestet ◽  
Richard Emsley ◽  
Sabine Landau
Keyword(s):  
Least Squares ◽  
Mediation Analysis ◽  
Indirect Effects ◽  
Ordinary Least Squares ◽  
Least Squares Estimator ◽  
Two Stage ◽  
Causal Mediation ◽  
Causal Mediation Analysis ◽  
Two Stage Least Squares ◽  
The Impact

Causal mediation analysis aims to estimate natural direct and natural indirect effects under clearly specified assumptions. Traditional mediation analysis based on Ordinary Least Squares assumes an absence of unmeasured causes to the putative mediator and outcome. When these assumptions cannot be justified, instrumental variable estimators can be used in order to produce an asymptotically unbiased estimator of the mediator-outcome link, commonly referred to as a Two-Stage Least Squares estimator. Such bias removal, however, comes at the cost of variance inflation. A Semi-Parametric Stein-Like estimator has been proposed in the literature that strikes a natural trade-off between the unbiasedness of the Two-Stage Least Squares procedure and the relatively small variance of the Ordinary Least Squares estimator. The Semi-Parametric Stein-Like estimator has the advantage of allowing for a direct estimation of its shrinkage parameter. In this paper, we demonstrate how this Stein-like estimator can be implemented in the context of the estimation of natural direct and natural indirect effects of treatments in randomized controlled trials. The performance of the competing methods is studied in a simulation study, in which both the strength of hidden confounding and the strength of the instruments are independently varied. These considerations are motivated by a trial in mental health, evaluating the impact of a primary care-based intervention to reduce depression in the elderly.

scholarly journals Initial coin offerings and the cryptocurrency hype - the moderating role of exogenous and endogenous signals

2021 ◽  
Author(s):  
Ferdinand Thies ◽  
Sören Wallbach ◽  
Michael Wessel ◽  
Markus Besler ◽  
Alexander Benlian
Keyword(s):  
Least Squares ◽  
Ordinary Least Squares ◽  
High Quality ◽  
Academic Interest ◽  
Entrepreneurial Ventures ◽  
Asset Class ◽  
The Impact ◽  
Moderating Role ◽  
Initial Coin Offerings

AbstractInitial coin offerings (ICOs) have recently emerged as a new financing instrument for entrepreneurial ventures, spurring economic and academic interest. Nevertheless, the impact of exogenous and endogenous signals on the performance of ICOs as well as the effects of the cryptocurrency hype and subsequent downfall of Bitcoin between 2016 and 2019 remain underexplored. We applied ordinary least squares (OLS) regressions based on a dataset containing 1597 ICOs that covers almost 2.5 years. The results show that exogenous and endogenous signals have a significant effect on the funds raised in ICOs. We also find that the Bitcoin price heavily drives the performance of ICOs. However, this hype effect is moderated, as high-quality ICOs are not pegged to these price developments. Revealing the interplay between hypes and signals in the ICO’s asset class should broaden the discussion of this emerging digital phenomenon.

DOES TRUST PROMOTE ENTREPRENEURSHIP IN A DEVELOPING COUNTRY?

2018 ◽  
Vol 63 (05) ◽  
pp. 1385-1403 ◽  
Author(s):  
KITAE SOHN ◽  
ILLOONG KWON
Keyword(s):  
Developing Countries ◽  
Property Rights ◽  
Least Squares ◽  
Developing Country ◽  
Ordinary Least Squares ◽  
Community Level ◽  
Two Stage ◽  
The Us ◽  
The Difference ◽  
The Individual

Trust was found to promote entrepreneurship in the US. We investigated whether this was true in a developing country, Indonesia. We failed to replicate this; this failure was true whether trust was estimated at the individual or community level or whether ordinary least squares (OLS) or two stage least squares (2SLS) was employed. We reconciled the difference between our results and those for the US by arguing that the weak enforcement of property rights in developing countries and the consequent hold-up problem make it more efficient for entrepreneurs to produce generic goods than relationship-specific goods—producing generic goods does not depend on trust.

scholarly journals The Impact of Industrialization, Trade Openness, Financial Development, and Energy Consumption on Economic Growth in Indonesia

Economies ◽  
2021 ◽  
Vol 9 (4) ◽  
pp. 174
Author(s):  
Khalid Eltayeb Elfaki ◽  
Rossanto Dwi Handoyo ◽  
Kabiru Hannafi Ibrahim
Keyword(s):  
Economic Growth ◽  
Energy Consumption ◽  
Least Squares ◽  
Financial Development ◽  
Correction Term ◽  
Trade Openness ◽  
Ordinary Least Squares ◽  
Distributed Lag ◽  
Long Run ◽  
The Impact

This study aimed to scrutinize the impact of financial development, energy consumption, industrialization, and trade openness on economic growth in Indonesia over the period 1984–2018. To do so, the study employed the autoregressive distributed lag (ARDL) model to estimate the long-run and short-run nexus among the variables. Furthermore, fully modified ordinary least squares (FMOLS), dynamic least squares (DOLS), and canonical cointegrating regression (CCR) were used for a more robust examination of the empirical findings. The result of cointegration confirms the presence of cointegration among the variables. Findings from the ARDL indicate that industrialization, energy consumption, and financial development (measured by domestic credit) positively influence economic growth in the long run. However, financial development (measured by money supply) and trade openness demonstrate a negative effect on economic growth. The positive nexus among industrialization, financial development, energy consumption, and economic growth explains that these variables were stimulating growth in Indonesia. The error correction term indicates a 68% annual adjustment from any deviation in the previous period’s long-run equilibrium economic growth. These findings provide a strong testimony that industrialization and financial development are key to sustained long-run economic growth in Indonesia.

scholarly journals Determinants of Credit Risk in the Banking Sector of Ghana: A Panel Co-integration Approach

2020 ◽  
Vol 5 (4) ◽  
Author(s):  
Thomas Appiah ◽  
Frank Bisiw
Keyword(s):  
Credit Risk ◽  
Least Squares ◽  
Banking Sector ◽  
Ordinary Least Squares ◽  
Capital Adequacy ◽  
High Rate ◽  
Long Run ◽  
Integration Techniques ◽  
The Government ◽  
The Impact

The economic development of any nation hinges on the health of its financial system. In recent years, the health of the Ghanaian Banking sector has been affected severely as a result of high levels of non-performing loans (NPLs), which has been identified as a major threat to the overall profitability and survival of banks. To minimize the impact of NPLs on the financial sector, key stakeholders such as the government, bank officials and regulators are working hard in that regard. However, any policy response aimed at dealing with the high rate of non-performing loans first requires the understanding of the underlying determinants of NPLs. Against this backdrop, this paper apply panel co-integration techniques to investigate the determinants of credit risk (NPLs) in the banking sector of Ghana.  We use NPL as a proxy to measure credit risk and assess how it is influenced by macroeconomic and bank-specific factors. A balanced panel data of 16 universal banks in Ghana from 2010 to 2016 has been analyzed using Panel co-integration techniques such as Fully Modified Ordinary Least Squares (FMOLS) and Dynamic Ordinary Least Squares (DOLS). Our result shows that growth in the economy, measured by Gross Domestic Product (GDP) has significant influence on the NPLs of banks in the long-run. The results further revealed that capital adequacy, profitability and liquidity of banks are significant predictors of NPLs. However, our results suggest that bank size, inflation and interest rate have statistically insignificant influence on the NPLs of Ghanaian banks. The study recommend, among others, that whereas it is important for government and policymakers to work to improve macroeconomic outcomes, banks should also improve their capital adequacy, profitability, and efficiency position as these bank-specific interventions could significantly improve credit quality and minimize NPLs.

The impact of gender inequality in education and employment on economic performance in Turkey: evidence from a cointegration approach

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Ömer Esen ◽  
Gamze Yıldız Seren
Keyword(s):  
Least Squares ◽  
Economic Performance ◽  
Ordinary Least Squares ◽  
Estimation Methods ◽  
Content Type ◽  
Cointegration Tests ◽  
Johansen Cointegration ◽  
Education And Employment ◽  
The Impact

PurposeThis study aims to empirically examine the impact of gender-based inequalities in both education and employment on economic performance using the dataset of Turkey for the period 1975–2018.Design/methodology/approachThis study employs Johansen cointegration tests to analyze the existence of a long-term relation among variables. Furthermore, dynamic ordinary least squares (DOLS) and fully modified ordinary least squares (FMOLS) estimation methods are performed to determine the long-run coefficients.FindingsThe findings from the Johansen cointegration analysis confirm that there is a long-term cointegration relation between variables. Moreover, DOLS and FMOLS results reveal that improvements in gender equality in both education and employment have a strong and significant impact on real gross domestic product (GDP) per capita in the long term.Originality/valueThe authors expect that this study will make remarkable contributions to the future academic studies and policy implementation, as it examines the relation among the variables by including the school life expectancy from primary to tertiary based on the gender parity index (GPI), the gross enrollment ratio from primary to tertiary based on GPI and the ratio of female to male labor force participation (FMLFP) rate.

scholarly journals The human capital–economic growth nexus in SSA countries: what can strengthen the relationship?

2020 ◽  
Vol 47 (9) ◽  
pp. 1143-1159
Author(s):  
Roseline Tapuwa Karambakuwa ◽  
Ronney Ncwadi ◽  
Andrew Phiri
Keyword(s):  
Economic Growth ◽  
Human Capital ◽  
Least Squares ◽  
Ordinary Least Squares ◽  
African Countries ◽  
Content Type ◽  
Econometric Approach ◽  
The Impact ◽  
The Relationship ◽  
Ssa Countries

PurposeThe purpose of this study is to examine the impact of human capital on economic growth for a selected sample of nine SSA countries between 1980 and 2014 using a panel econometric approach.Design/methodology/approachThe authors estimate a log-linearized endogenous using the fully modified ordinary least squares (FMOLS) and the dynamic ordinary least squares (POLS) applied to our panel data time series.FindingsThe empirical analysis shows an insignificant effect of human capital on economic growth for our selected sample. These findings remain unchanged even after adding interactive terms to human capital, which are representatives of government spending as well as foreign direct investment. Nevertheless, the authors establish a positive and significant effect of the interactive term between urbanization and human capital on economic growth.Practical implicationsThe results emphasize the need for African policymakers to develop urbanized, “smart”, technologically driven cities within the SSA region as a platform toward strengthening the impact of human capital-economic growth relationship.Originality/valueThis study becomes the first in the literature to validate the human capital–urbanization–growth relationship for African countries.

Mandatory CSR expenditure and stock return

2020 ◽  
Vol 28 (6) ◽  
pp. 951-975
Author(s):  
Asit Bhattacharyya ◽  
Md Lutfur Rahman
Keyword(s):  
Least Squares ◽  
Stock Returns ◽  
Least Squares Method ◽  
Negative Impact ◽  
Generalized Method Of Moments ◽  
Ordinary Least Squares ◽  
Reverse Causality ◽  
Endogeneity Bias ◽  
Content Type ◽  
The Impact

Purpose India has mandated corporate social responsibility (CSR) expenditure under Section 135 of the Indian Companies Act, 2013 – the first national jurisdiction to do so. The purpose of this paper is to examine the impact of mandated CSR expenditure on firms’ stock returns by using actual CSR spending data, whereas the previous studies mostly focus on voluntary CSR proxied by CSR scores. Design/methodology/approach The authors estimate their baseline regression by using ordinary least squares(OLS) method. Although the baseline regression involving CSR expenditure and stock returns using ordinary least squares method are estimated, endogeneity and reverse causality biases are addressed by using two-stage least squares and generalized method of moments approaches. These approaches contribute mitigating endogeneity bias and biases associated with unobserved heterogeneity and simultaneity. Findings The findings document that mandatory CSR expenditure has a negative impact on firms’ stock returns which supports the “shareholders” expense’ view. This result remain robust after controlling for endogeneity bias and the use of both standard and robust test statistics. The authors however observe that this result holds for the firms with actual CSR expenditure equal to the mandated amount but does not hold for the firms with actual CSR expenditure greater than the mandated amount. Therefore, the authors provide evidence that CSR expenditure’s impact on stock returns depends on whether firms simply comply the regulation or voluntarily chose an amount of CSR expenditure above the mandated amount. Originality/value The primary contribution is to present a valid and robust evidence of negative effect of mandated CSR spending on firms’ stock returns when the mandatory CSR spending rule is already in place. This study contributes by examining the impact of mandated CSR spending on stock during post-implementation period (2015-2017), whereas other studies by Dharampala and Khanna (2018); Kapoor and Dhamija (2017); and Mukherjee et al. (2018) mainly examined the impact of legislation on Indian CSR. The authors use mandated actual CSR expenditure, whereas previous studies mostly focus on voluntary CSR proxied by CSR scores.

The Analysis and Forecast of Housing Requirements: Tenure Switching in a Local Housing Market

1981 ◽  
Vol 13 (1) ◽  
pp. 57-64
Author(s):  
R K Wilkinson ◽  
I A Law
Keyword(s):  
Regression Analysis ◽  
Housing Market ◽  
Least Squares ◽  
Multiple Regression Analysis ◽  
Multiple Regression ◽  
Local Authority ◽  
Ordinary Least Squares ◽  
Reduced Form ◽  
Annual Number ◽  
Demand Structure

Data collected by a local authority housing department are analysed in order to test a model forecasting the annual number of households moving out into owner occupation. The reduced form of the model is estimated by means of ordinary least squares multiple regression analysis, and the demand structure is estimated by means of two-stage least squares for comparison. Despite problems of autocorrelation the results are sensible and suggest that further work on these lines is likely to be worthwhile not least for a local authority involved in planning the housing requirements of a region.

Sign in / Sign up

Export Citation Format

Share Document