Standard OLS Estimation Research Articles

Improving Nigeria’s Inflation Forecast with Oil Price: The Role of Estimators

In this study, we propose a supply-side augmented Phillips curve for an oil dependent (Nigerian) economy. We argue for the role of oil price as a good proxy for the supply side of inflation given the structure of the Nigerian economy, which essentially relies on oil revenue. Thus, we compare the forecast results of the oil-based augmented Phillips curve with the traditional variant, as well as time series models such as ARIMA and ARFIMA. We also test for any probable asymmetric response of Nigeria’s inflation forecast to oil price changes. The forecast analyses are conducted for both in-sample and out-of-sample periods using alternative forecast measures. We employ the estimators of Lewellen (J Financ Econ 74:209–235, 2004) and Westerlund and Narayan (J Financ Econ 13(2):342–375, 2015) in order to account for relevant statistical properties of the predictors, and their results are compared with the standard OLS estimator. In addition, we follow the extended version of Westerlund and Narayan, constructed into a linear multi-predictive form by Makin et al. (J Int Money Financ 40:63–78, 2014) and Salisu et al. (Energy Econ, 2018), and a nonlinear (asymmetric) multi-predictive model by Salisu and Isah (Econ Model, 2018). We find that the augmented (oil-based) Phillips curve outperforms its traditional version, as well as time series models for both forecast samples. However, oil price asymmetric effects become evident when large samples are employed. Also, we find that the choice of estimator does matter for accurate inflation forecasts and by implication, accounting for the salient features of the predictors, if they exist, has implications on the forecast performance. Our results are robust to alternative oil price proxies and forecast measures.

Examining monetary policy reaction in the People’s Republic of China – a Markov switching policy index approach

This paper estimates a monetary policy rule for the People’s Republic of China (PRC) using a standard OLS estimation and a Markov switching model. As the People’s Bank of China (PBOC) generally uses a battery of instruments in the conduct of its monetary policy, these models are estimated using a constructed monetary policy index (MPI) in place of the traditional interest rate. This allows for a better understanding of the role the PBOC has played in the PRC’s unprecedented economic growth and its relatively low inflation over the last twenty years. This paper will not only examine the unique characteristics of Chinese monetary policy but may also give a more general insight into the dynamics of monetary policy reactions in other emerging markets and economies in transition.

Asymmetric fuel price responses under heterogeneity

We explore the effect of cross-sectional aggregation of data on estimation and test of asymmetric retail fuel price responses to wholesale price shocks. The analysis is performed on data collected daily from individual fuel stations in the Spanish metropolitan areas of Madrid and Barcelona. While the standard OLS estimator is applied to an error correction model in the case of the aggregated time series, we use the mean group approaches developed by Pesaran and Smith (1995) and Pesaran (2006) to estimate the short- and long-run micro-relations under heterogeneity. We found remarkable differences between the results of estimations using aggregated and disaggregated data, which are highly robust to both datasets considered. Our findings could help to explain many of the results in the literature on this research topic. On the one hand, they suggest that the typical estimation with aggregated data clearly tends to overestimate the persistence of shocks. On the other hand, we show that aggregation may generate a loss of efficiency in econometric estimates that is sufficiently large to hide the existence of the “rockets and feathers” phenomenon.

Equilibrium vs. optimal city size: evidence from Italian cities

In this paper, the stylised assumption that one single ‘optimal’ city size exists for all cities – achieved when marginal location costs equal marginal location benefits – is abandoned, as well as the opposite view that each city operates on its own cost and production curves, defining a specific optimal size. Instead, this work maintains the comparability among cities and demonstrates that urban specificities in functions performed, quality of life, industrial diversity and social conflicts shift up and down the benefits and costs linked to pure physical size, leading to different ‘equilibrium’ sizes for cities. A model of equilibrium urban size is set up, and empirically estimated on a sample of 103 Italian cities with data at NUTS3 level. Empirical results verify the empirical model on the analysed sample; results hold both with standard OLS estimates as well as with the use of instrumental variables in order to correct for the possible endogeneity of some of the variables in the model. Differences between predicted and real city size are interpreted with good or bad governance, thereby suggesting future strategies for more efficient urban planning.

Conditional Returns to Shareholders of Bidding Firms: An Australian Study

This study examines the importance of the self-selection problem when evaluating returns to bidder firms around announcement events. Takeover announcements are not random because managers decide rationally whether to bid or not, this indicates announcements are timed; consequently, in the presence of sample selection problem, standard OLS estimates are biased. Using a conditional model the results indicate that after controlling for the self-selection bias effect, shareholders of bidder firms make normal returns. Results are robust and consistent with conventional economic theory. In sum failing to account for sample selection bias may lead to erroneous conclusions about a bidder’s true economic wealth effects around an announcement event.

Detecting Contagion Effects During The Subprime Crisis Using Different VAR Size Models: Comparison Between OLS And Bayesian Shrinkage

<span style="font-family: Times New Roman; font-size: small;"> </span><p style="margin: 0in 0.5in 0pt; text-align: justify; mso-pagination: none;" class="MsoNormal"><span style="color: black; font-size: 10pt; mso-ansi-language: EN-US;"><span style="font-family: Times New Roman;">This paper assesses the performance in terms of forecast and structural analysis of different sizes of VAR models while studying contagion effects during the subprime crisis. We compare VAR models of different sizes: a standard VAR estimated by OLS, and a MEDIUM and LARGE VARs estimated by a Bayesian shrinkage procedure. </span></span></p><span style="font-family: Times New Roman; font-size: small;"> </span><p style="margin: 0in 0.5in 0pt; text-align: justify; mso-pagination: none;" class="MsoNormal"><span style="color: black; font-size: 10pt; mso-ansi-language: EN-US;"><span style="font-family: Times New Roman;"> </span></span></p><span style="font-family: Times New Roman; font-size: small;"> </span><p style="background: white; margin: 0in 0.5in 0pt; text-align: justify; mso-pagination: none;" class="MsoNormal"><span style="color: black; font-size: 10pt; mso-ansi-language: EN-US;"><span style="font-family: Times New Roman;">Using a large database containing national macroeconomic, financial, and trade dynamic variables for 17 OECD countries, we study contagion effects on U.S key variables and OECD housing markets over the period of the subprime crisis.</span></span></p><span style="font-family: Times New Roman; font-size: small;"> </span><p style="background: white; margin: 0in 0.5in 0pt; text-align: justify; mso-pagination: none;" class="MsoNormal"><span style="color: black; font-size: 10pt; mso-ansi-language: EN-US;"><span style="font-family: Times New Roman;"> </span></span></p><span style="font-family: Times New Roman; font-size: small;"> </span><p style="background: white; margin: 0in 0.5in 0pt; text-align: justify; mso-pagination: none;" class="MsoNormal"><span style="color: black; font-size: 10pt; mso-ansi-language: EN-US;"><span style="font-family: Times New Roman;">Our paper unearths that the smallest specification outperforms the largest one, in terms of forecast accuracy and structural analysis. Our results reveal that the SMALL and MEDIUM VARs produce qualitatively similar results; however, the MEDIUM VAR permits us to obtain the responses of a large set of variables and so provides a more comprehensive picture of the effects of the subprime crisis. So, we conclude that the MEDIUM VAR is more successful at extracting pertinent information for a large data set. It appears that dealing with a LARGE VAR provides worst results.</span></span></p><span style="font-family: Times New Roman; font-size: small;"> </span><p style="background: white; margin: 0in 0.5in 0pt; text-align: justify; mso-pagination: none;" class="MsoNormal"><span style="color: black; font-size: 10pt; mso-ansi-language: EN-US;"><span style="font-family: Times New Roman;"> </span></span></p><span style="font-family: Times New Roman; font-size: small;"> </span><p style="margin: 0in 0.5in 0pt; text-align: justify; mso-pagination: none; mso-layout-grid-align: none;" class="MsoNormal"><span style="color: black; font-size: 10pt; mso-ansi-language: EN-US;"><span style="font-family: Times New Roman;">An interesting result that arises is that the standard OLS estimation is interesting; however, the Bayesian shrinkage is a useful tool, in the case of an international contagion study, allowing both a more comprehensive picture of the contagion effects as well as a more complete check of the empirical plausibility of the underlying specification.</span></span></p><span style="font-family: Times New Roman; font-size: small;"> </span>

Keeping up with the Joneses: Neighborhood effects in housing renovation

Despite widespread recognition that housing renovation is influenced by “neighborhood effects”, virtually all empirical studies have failed to identify a positive feedback effect between renovation activity and neighborhood quality. By explicitly modeling the spatial interdependence of households' renovation decisions and analyzing a detailed block-level data set, this study finds strong empirical evidence that endogenous neighborhood effects exist as expected. Moreover, by considering four different parameterizations of a “neighborhood set” and comparing the results of these spatial econometric models with a standard OLS estimation, this paper provides insight into some common methodological issues encountered when modeling neighborhood effects.

Azar y fortuna en España: la Lotería Nacional, 1850–2000

This article examines the determinants of the Spanish National Lottery between 1850 and 2000. The sales data are from Gaceta de Madrid (1850–1920), Cuentas de Tesoreros (Sección de Hacienda) of Administración General Estado Archive (1939–1955) and the Annual Reports from LAE (1985–2000). The results provide a detailed examination of the socioeconomic and demographic characteristics of lottery players –income, educational levels, urbanisation, number of lottery retail outlets in the province, age and religion- that historically determine lottery ticket purchases in Spain. This study uses standard OLS estimation and models that account spatial dependence between a province and its neighbors.

The effects of over-education on earnings in the graduate labour market

This paper uses a new survey of graduates from one large civil university in the UK to examine the determinants of over-education and its subsequent impact on labour market earnings. Multiple measurements of over-education were collected to assess the effect of measurement error on the estimated pay penalty associated with over-education. Panel estimates suggest that the upward bias in standard OLS estimates is offset by an equal downward bias resulting from measurement error.

Zvi Griliches' Contribution to the Theory of Human Capital

The paper discusses Zvi Griliches' contribution to the estimation of the earning function. The topic was the central theme of Griliches' research agenda during the 70s. Griliches played a major role in the ability- schooling controversy of the time. He was instrumental in repelling the attack of the revisionists' on the Theory of Human Capital, and the claim that the schooling effect in the earning function is merely an artifact of the true ability' and family background' effects. Griliches lacked at the time the proper data to prove unequivocally that the ability bias plays only a minor role in the estimation of the rate of return to schooling. He was, however, able to show that the seemingly foolproof evidence of his opponents suffers from serious biases due to the endogeneity and the measurement errors in the schooling variable. His assertion that the standard OLS estimator is biased downward, rather than upward, has been shown true by future research.

Performance Thresholds in Managerial Incentive Contracts*

Performance thresholds are commonly used in executive compensation contracts. We examine the contractual nonlinearity associated with performance thresholds and its incentive implications. Incorporating a performance threshold into a standard principal‐agent model of a linear contract, we show that pay schemes using a performance threshold are optimal. By truncating a linear scheme at poor performance, the threshold mitigates agency costs associated with the downside risk of production. Examining CEO compensation data, we find evidence of the role of performance thresholds. As a consequence of under‐threshold performance, the tobit estimator is shown to increase pay‐performance sensitivity, notably improving upon the standard OLS estimator.

An Alternative Approach to the Specification of Structural Transition Functions

This paper forwards an alternative approach of characterizing the adjustment trajectory by a cumulative probability distribution function. Estimates of Wilton's (1975) automobile import equation over the period 1984-1973 are obtained in the case where a binomial CDF is used. Confirming Wilton's results, the use of transition functions to characterize coefficient adjustment between two regimes leads to substantial improvements over standard OLS estimation. Also the CDF transition estimates compare very favourably with the polynomial results, at the same time yielding more reliable coefficient transition paths and an improved framework for more precise hypothesis testing of structural change. (This abstract was borrowed from another version of this item.)