AgEcon Search
http://ageconsearch.umn.edu
AgEcon Search latest documentsenThu, 21 Jun 2018 20:17:37 GMTInvenio UNKNOWN-dirtyaesearch@umn.edu360663125http://ageconsearch.umn.edu/img/site_logo_rss.pngAgEcon Search
http://ageconsearch.umn.edu
Search Search this site:p
http://ageconsearch.umn.edu/search
Numerical Distribution Functions of Fractional Unit Root and Cointegration Tests
http://ageconsearch.umn.edu/record/273739
We calculate, by simulations, numerical asymptotic distribution functions of likelihood ratio tests for fractional unit roots and cointegration rank. Because these distributions depend on a real-valued parameter b which must be estimated, simple tabulation is not feasible. Partly due to the presence of this parameter, the choice of model specification for the response surface regressions used to obtain the numerical distribution functions is more involved than is usually the case. We deal with model uncertainty by model averaging rather than by model selection. We make available a computer program which, given the dimension of the problem, q, and a value of b, provides either a set of critical values or the asymptotic P value for any value of the likelihood ratio statistic.MacKinnon, James G.Wed, 20 Jun 2018 14:36:07 GMThttp://ageconsearch.umn.edu/record/2737392010-07Covariance-based orthogonality tests for regressors with unknown persistence
http://ageconsearch.umn.edu/record/273598
This paper develops a new test of orthogonality based on a zero restriction on the covariance between the dependent variable and the predictor. The test provides a useful alternative to regression-based tests when conditioning variables have roots close or equal to unity. In this case standard predictive regression tests can suffer from well-documented size distortion. Moreover, under the alternative hypothesis, they force the dependent variable to share the same order of integration as the predictor, whereas in practice the dependent variable often appears stationary while the predictor may be near-nonstationary. By contrast, the new test does not enforce the same orders of integration and is therefore capable of detecting alternatives to orthogonality that are excluded by the standard predictive regression model. Moreover, the test statistic has a standard normal limit distribution for both unit root and local-to-unity conditioning variables, without prior knowledge of the local-to-unity parameter. If the conditioning variable is stationary, the test remains conservative and consistent. Thus the new test requires neither size correction nor unit root pre-test. Simulations suggest good small sample performance. As an empirical application, we test for the predictability of stock returns using two persistent predictors, the dividendprice- ratio and short-term interest rate.Maynard, AlexWed, 13 Jun 2018 17:30:18 GMThttp://ageconsearch.umn.edu/record/2735982007-02Which factors influence farmers’ intentions to adopt nutrient management planning?
http://ageconsearch.umn.edu/record/273498
Nutrients not converted to agricultural products are at risk of being lost to the environment and can contribute to environmental degradation. The adoption of nutrient management planning (NMP) can lead to win-win outcomes in terms of both improving productivity and reducing the environmental impact of farming, yet adoption remains below expectations globally. Few studies specifically focus on the adoption of NMP and the majority overlook psychological factors in their analysis. This study examines the factors which influence Irish farmers’ intention to adopt NMP as defined by intention to apply fertiliser on the basis of soil test results. An expanded version of the theory of planned behaviour is used as a framework for analysis. The influence of policy is also accounted for by this study which requires certain farmers in Ireland to adopt soil testing on a mandatory basis. The results for the national sample (n=1009) show that attitudes, subjective norms (social pressure), perceived behavioural control (ease/difficulty) and perceived resources are significant and positively associated with farmers’ intentions. In terms of the voluntary sample (n=587), only attitude, perceived behavioural control and perceived resources are significantly and positively associated with farmers’ intentions. Whereas, for the mandatory sample (n=422), subjective norms, perceived behavioural control and perceived resources are significantly and correlated in a positive direction with intentions. A number of farm and farmer characteristics were also significantly associated with intentions. We argue that psychological issues must be addressed by initiatives designed to encourage further uptake of NMP. Furthermore, initiatives must account for the heterogeneity in the factors which influence farmers to adopt, which may be influenced by motivations for adopting soil testing, such as voluntary or mandatory purposes. Daxini, AmarMon, 11 Jun 2018 20:01:57 GMThttp://ageconsearch.umn.edu/record/2734982018-04-17Which factors influence farmers’ intentions to adopt nutrient management planning?
http://ageconsearch.umn.edu/record/273494
Nutrients not converted to agricultural products are at risk of being lost to the environment and can contribute to environmental degradation. The adoption of nutrient management planning (NMP) can lead to win-win outcomes in terms of both improving productivity and reducing the environmental impact of farming, yet adoption remains below expectations globally. Few studies specifically focus on the adoption of NMP and the majority overlook psychological factors in their analysis. This study examines the factors which influence Irish farmers’ intention to adopt NMP as defined by intention to apply fertiliser on the basis of soil test results. An expanded version of the theory of planned behaviour is used as a framework for analysis. The influence of policy is also accounted for by this study which requires certain farmers in Ireland to adopt soil testing on a mandatory basis. The results for the national sample (n=1009) show that attitudes, subjective norms (social pressure), perceived behavioural control (ease/difficulty) and perceived resources are significant and positively associated with farmers’ intentions. In terms of the voluntary sample (n=587), only attitude, perceived behavioural control and perceived resources are significantly and positively associated with farmers’ intentions. Whereas, for the mandatory sample (n=422), subjective norms, perceived behavioural control and perceived resources are significantly and correlated in a positive direction with intentions. A number of farm and farmer characteristics were also significantly associated with intentions. We argue that psychological issues must be addressed by initiatives designed to encourage further uptake of NMP. Furthermore, initiatives must account for the heterogeneity in the factors which influence farmers to adopt, which may be influenced by motivations for adopting soil testing, such as voluntary or mandatory purposes. Daxini, AmarMon, 11 Jun 2018 19:41:53 GMThttp://ageconsearch.umn.edu/record/2734942018-04-17Welfare Effect of Urea Deep Placement (UDP) Technology Adoption among Smallholder Rice Farmers in Kwara State, Nigeria – Analysis of a Randomized Control Trial Experiment
http://ageconsearch.umn.edu/record/273493
We analyse the effects of adoption of urea deep placement (UDP) technology on household welfare using household data collected through a randomized control trial experiment conducted among rice farming households in Kwara State, Nigeria. In order to adjust for unequal sampling fractions and correct for possible misspecification and selection bias in the effects model, we estimate a treatment effect model using the doubly robust inverse probability weighted regression adjustments. The results show that variable that enhances intra-household joint decision-making on farm input use and wealth indicators encourage UDP adoption while physiological risks tend to discourage adoption. Differential use of food consumption coping strategies exists between adopters and non-adopters along less severe but not severe strategies. We find that UDP technology can significantly increase food security, measured as food consumption coping strategy index, among adopting households. This implies that the technology is welfare increasing, and its adoption could be discouraged by health and production – related risks interventions.Edeh, HyacinthMon, 11 Jun 2018 19:35:50 GMThttp://ageconsearch.umn.edu/record/2734932018-04-17Heterogeneous Credit Constraints and Smallholder Farming in Senegal
http://ageconsearch.umn.edu/record/273491
Credit constraints are among key challenges to unlocking the great economic and social potentials of small farm agriculture in sub-Saharan Africa. This research sets out to analyze the extent to which farmers are credit-constrained, the underlying generating mechanisms, and how financial inclusion through the reduction or elimination of credit constraints would benefit smallholders in the specific agro-ecological region of the Senegal River Valley. So far the literature tends to focus on credit applicants when defining access to credit dummy, ignoring in the process the vast majority of farmers who stay out of the market. Instead, the paper recognizes that credit constraints come in different forms to the extent that they translate into market entry barriers at the pre-application stage (ex-ante) or contribute to deteriorate the credit profile at the post-application stage (ex-post). Farm-level data are used, and a model that controls for both endogeneity and farmers’ self-selection into the credit market is developed. The results suggest that credit constraints, mostly originated from high transaction costs and high risk, are harming farmers’ performance, and access to credit leads to increased yields and labor productivity. The extent of the gains depends on the stage at which the constraints manifest themselves, as well as the performance indicator and the reference group. These results suggest various policy options to be considered in order to unlock the economic and social potentials associated with financial inclusion in the farming sector.Seck, AbdoulayeMon, 11 Jun 2018 19:29:14 GMThttp://ageconsearch.umn.edu/record/2734912018-04-17The Decision to Adopt Organic Practices in Malaysia; a Mix-method Approach
http://ageconsearch.umn.edu/record/273485
Organic agriculture is being promoted globally and is considered to have beneficial impacts on sustainable development. Aside from the importance of organic agriculture for human health and the environment, organic farming also indirectly contributes to income generation, development of new technologies and indigenous knowledge, as well as supporting rural development (Darnhofer, 2005; Scialabba, 2000). In Malaysia, people have recently become more aware of the potential advantages of organic food; however, the adoption rate among farmers is still very low and the number of certified farmers remains small. In fact, the demand from consumers for organic products continues to increase and cannot be met by local producers, hence relying on imports from other countries. In this context, this paper aims to investigate the factors that influence the farmer’s decision to adopt organic farming practices. This study was carried out in all four regions of Peninsular Malaysia (Northern, Central, Southern, and East Coast) which also represent the four highest number of organic farmers that have been listed by the Department of Agriculture (DoA) Malaysia. Logistic regression analysis was estimated to ascertain the factors. The paper intends to explain the decision to adopt or not adopt organic farming with respect to a variety of socio-economic, social and attitudinal factors. The analysis was done using both qualitative and quantitative methods, towards a more comprehensive understanding of the overall situation of organic farming practices in Malaysia. The results indicate that environmental attitudes is the most influential factor, followed by information sharing attitudes, as well as land ownership. These observations can contribute to national policy development in the context of future sustainable development. Mohamed Haris, Nur BahiahMon, 11 Jun 2018 18:53:45 GMThttp://ageconsearch.umn.edu/record/2734852018-04-16PROTECTED GEOGRAPHICAL INDICATION RECOGNITION AND WILLINGNESS TO PAY: A CASE OF GROJEC APPLE
http://ageconsearch.umn.edu/record/273276
The Grojec region of Poland is an important region for apple production and accounts for 40 percent of domestic apple production. Apple growers from the region made an attempt to strengthen their competitive position through registering their apples as Protected Geographical Indication (PGI) products. The European Commission’s PGI allows food producers to obtain market recognition and a premium price for their products. Although the Grojec Apple received PGI registration in 2011, little has been done to promote apples with the PGI label. Two important research questions are addressed: 1) Does the Polish market recognize Grojec Apple PGI, and 2) Does the market value Grojec Apple PGI? Logit and regression models are estimated using survey data collected during an International MBA in Agribusiness and Commerce study week in Warsaw. Only 22% of consumers recognize Grojec Apple PGI. Yet, 70% of consumers indicate they are willing to pay more for the product and their average willingness to pay (WTP) premium is 32%. Results indicate use of the PGI label may be effective in improving sales and profit margins for Grojec Apple producers and their affiliated cooperatives. Older consumers are more likely to indicate a WTP premium. Males, smaller households, and consumers less sensitive to apple price indicate a higher WTP premium. An advertising campaign promoting Grojec Apple PGI as a better product may be effective at increasing consumer likelihood to pay more and WTP premium. Although “Grojec” is already familiar to most consumers in central Poland as a region for apples, a Grojec Apple with PGI label would assure consumers they are purchasing apples from the Grojec region and the apples are high quality.Bruce L., AhrendsenWed, 06 Jun 2018 15:16:04 GMThttp://ageconsearch.umn.edu/record/273276urn:ISSN:1789-221X2017-12-31ECONOMIC ISSUES OF DUCK PRODUCTION: A CASE STUDY FROM HUNGARY
http://ageconsearch.umn.edu/record/273274
The Hungarian waterfowl sector is characterised by export orientation, as 55-57% of the revenue comes from exports, so its importance is high in the national economy. The production of slaughter animals in the duck sector has doubled in the last decade. The objective of the study is to examine production parameters, as well as the cost and profit situation of broiler duck production and to reveal the correlations between the factors with a case study through the example of a Hungarian company. The production parameters and cost data of the investigated farm (2014-2016, 96 production cycles) were analysed using descriptive statistical methods, correlation and regression analysis. The results show that the average cost of the duck produced in intensive, closed farming system was between 72.6 and 101.7 eurocent kg-1. The most significant cost items were feed (52-63%) and chicken cost (14-19%). The sales price decreased from 112.9 eurocent kg-1 to 98.4 eurocent kg-1 during the examined period, resulting in a profit from -3.3 to 25.7 eurocent kg-1, and overall profitability was decreasing. The study also revealed that there was no correlation between average cost and final bodyweight, while the correlation between average cost and reared period was weak. At the same time, the relationship between average cost and average daily weight gain, mortality, feed conversion ratio was moderate. In addition, the European Production Efficiency Factor (EPEF) can be adapted to the duck sector as strong, positive relationship can be scientifically verified between the indicator and average cost. There is a close correlation between the sold live weight per m2 and the amount of feed used per m2, as well as between the final bodyweight and the amount of feed used to rear a duck, while the correlation between average cost and the sold live weight per m2 is weak.Szilvia, MolnárWed, 06 Jun 2018 15:05:50 GMThttp://ageconsearch.umn.edu/record/273274urn:ISSN:1789-221X2017-12-31Semiparametric Specification Testing
http://ageconsearch.umn.edu/record/273180
Delgado, Miguel A.Wed, 30 May 2018 16:20:17 GMThttp://ageconsearch.umn.edu/record/2731801990-03Direct payments and competitiveness. Assessing redistributive effects of internal convergence in Italy
http://ageconsearch.umn.edu/record/271955
The objective of the paper is twofold. Firstly, investigating the relationship between competitiveness, measured by technical efficiency, and direct payments of a sample of Italian farms prior to the application of the 2014-2020 CAP reform. Secondly, evaluating possible implications of alternative scenarios about distribution of direct payments on technical efficiency. To these aims, a data envelopment analysis approach in conjunction with a double bootstrapped left-truncated regression model are adopted. Results indicate that direct payments are negatively associated with technical efficiency. Moreover, they show that redistribution of policy subsidies especially towards more inefficient farms provokes a decrease in overall technical efficiency. Bonfiglio, AndreaSun, 22 Apr 2018 11:43:24 GMThttp://ageconsearch.umn.edu/record/2719552018-04-26Disadvantage and Economic Viability: Evidence from Scottish Upland Farming
http://ageconsearch.umn.edu/record/271954
With the UK’s withdrawal from the European Union and calls for establishing a new agricultural policy, it is pertinent to examine the role that disadvantage plays within support for farm incomes. We matched farm business survey data with land capability and other biophysical data to understand the impact of biophysical and distance based disadvantage on viability. We derive five indicators of viability, ranging from vulnerable to robust and resilient. An ordinal logistic regression then measures the effect of biophysical and distance based disadvantage on viability. We find biophysical criteria affects viability, but other factors such as attitudes and farm structural decision making has a positive influence of viability status. Barnes, AndrewSun, 22 Apr 2018 11:40:06 GMThttp://ageconsearch.umn.edu/record/2719542018-04-26Estimating and modeling relative survival
http://ageconsearch.umn.edu/record/271694
When estimating patient survival using data collected by population-based cancer registries, it is common to estimate net survival in a relative-survival framework. Net survival can be estimated using the relative-survival ratio, which is the ratio of the observed survival of the patients (where all deaths are considered events) to the expected survival of a comparable group from the general population. In this article, we describe a command, strs, for life-table estimation of relative survival. We discuss three methods for estimating expected survival, as well as the cohort, period, and hybrid approaches for estimating relative survival. We also implement a life-table version of the Pohar Perme (2012, Biometrics 68: 113–120) estimator of net survival, and we describe two methods for age standardization. We also explain how, in addition to net probabilities of death, crude probabilities of death due to cancer and due to other causes can be estimated using the method of Cronin and Feuer (2000, Statistics in Medicine 19: 1729–1740). We conclude this article with discussion and examples of modeling excess mortality using various approaches, including the full-likelihood approach (using the ml command) and Poisson regression (using the glm command with a user-specified link function). Dickman, Paul W.Wed, 18 Apr 2018 14:18:19 GMThttp://ageconsearch.umn.edu/record/271694urn:ISSN:1536-86342015Impact of the SO Threshold on the Statistics of Economic Variables for the Swiss Agricultural Sector
http://ageconsearch.umn.edu/record/270170
In future, standard output (SO) will be the economic variable used to define the target population from which the Swiss FADN sample is drawn. This study assesses the impact of the SO threshold on key economic variables at national level. The analysis demonstrates that raising the SO threshold will lead to higher average values of key economic variables such as work income per family labour unit. This result is confirmed by two entirely different approaches, the first of which takes the FADN data into account, and the second of which considers census data supplemented by imputed economic variables. In Zukunft wird die Grundgesamtheit, aus welcher die Stichprobe des schweizerischen Informationsnetzes Landwirtschaftlicher Buchführungen (INLB) gezogen wird, mithilfe des Standardoutputs (SO) definiert. Die vorliegende Studie untersucht den Einfluss des SOSchwellenwertes auf den nationalen Mittelwert einiger wichtiger ökonomischer Variablen. Die Analyse zeigt, dass eine Zunahme der SO-Schwellenwerte zur Abgrenzung der Grundgesamtheit zu einer Erhöhung der Mittelwerte ökonomischer Parameter (z.B. Arbeitsverdienst pro Familienarbeitskraft) führt. Dieses Resultat wird durch zwei unterschiedliche Vorgehensweisen gestützt, wobei die erste auf den Daten des INLB beruht, während die zweite auf Daten der landwirtschaftlichen Strukturdatenerhebung und interpolierten ökonomischen Parametern basiert.Roesch, Andreas Thu, 29 Mar 2018 09:47:52 GMThttp://ageconsearch.umn.edu/record/270170urn:ISSN:0002-11212015-03-01Are Central Banks following a linear or nonlinear (augmented) Taylor rule?
http://ageconsearch.umn.edu/record/269883
The Taylor rule establishes a simple linear relation between the interest rate, inflation and output gap. However, this relation may not be so simple. To get a deeper understanding of central banks’ behaviour, this paper asks whether central banks are indeed following a linear Taylor rule or, instead, a nonlinear rule. At the same time, it also analyses whether that rule can be augmented with a financial conditions index containing information from some asset prices and financial variables. A forwardlooking monetary policy reaction function is employed in the estimation of the linear and nonlinear models. A smooth transition model is used to estimate the nonlinear rule. The results indicate that the European Central Bank and the Bank of England tend to follow a nonlinear Taylor rule, but not the Federal Reserve of the United States. In particular, those two central banks tend to react to inflation only when inflation is above or outside their targets. Moreover, our evidence suggests that the European Central Bank is targeting financial conditions, contrary to the other two central banks. This lack of attention to the financial conditions might have made the United States and the United Kingdom more vulnerable to the recent credit crunch than the Eurozone. Castro, VitorMon, 26 Mar 2018 13:38:50 GMThttp://ageconsearch.umn.edu/record/2698832008Are immigrants so stuck to the floor that the ceiling is irrelevant?
http://ageconsearch.umn.edu/record/269787
In this paper, the immigrant-native wage di§erential is explained through quantile regression estimations. Using repeated cross-sections of the British Labour Force Survey from 1993-2005, we analyse the returns to covariates across the conditional earnings distribution. We estimate a pooled model with an immigrant dummy and separate models for immigrants and natives of the UK. Our results show that the positive wage gap in favour of immigrants is attributed to those at higher quantiles. Returns to education and experience vary wider for natives than for immigrants. We decompose the wage gap in the Blinder-Oaxaca framework and apply quantile regression techniques to see if immigrants simply have more viable labour market characteristics than natives or if there is a preference for immigrant workers (reverse discrimination). Our Öndings suggest immigrants should actually be earning more and there is su¢ cient evidence of discrimination. This Önding is, however, not symmetric across the conditional wage distribution and immigrants at the bottom face more discrimination than those at the top.Hunt, PriscilliaThu, 22 Mar 2018 14:51:09 GMThttp://ageconsearch.umn.edu/record/2697872008Am I missing something? The effects of absence from class on student performance
http://ageconsearch.umn.edu/record/269769
We exploit a rich administrative panel data-set for cohorts of Economics students at a UK university in order to identify causal effects of class absence on student performance. We exploit the panel properties of the data to control for unobserved heterogeneity across students and hence for endogeneity between class absence and academic performance of students stemming from the likely influence of effort and ability on both absence and performance. Our estimations also exploit features of the data such as the random assignment of students to classes and information on the timetable of classes, which provides potential instruments in our identification strategy. Among other results we find, from a quantile regression specification, that there is a causal effect of absence on performance for students: missing class leads to poorer performance. There is evidence that this is particularly true for better-performing students, consistent with our hypothesis that effects of absence on performance are likely to vary with factors such as student ability.Arulampalam, WijiWed, 21 Mar 2018 21:35:45 GMThttp://ageconsearch.umn.edu/record/2697692007Quantile Graphical Models : Prediction and Conditional Independence with Applications to Financial Risk Management
http://ageconsearch.umn.edu/record/269321
We propose Quantile Graphical Models (QGMs) to characterize predictive and conditional independence relationships within a set of random variables of interest. This framework is intended to quantify the dependence in non-Gaussian settings which are ubiquitous in many econometric applications. We consider two distinct QGMs. First, Condition Independence QGMs characterize conditional independence at each quantile index revealing the distributional dependence structure. Second, Predictive QGMs characterize the best linear predictor under asymmetric loss functions. Under Gaussianity these notions essentially coincide but non-Gaussian settings lead us to different models as prediction and conditional independence are fundamentally different properties. Combined the models complement the methods based on normal and nonparanormal distributions that study mean predictability and use covariance and precision matrices for conditional independence. We also propose estimators for each QGMs. The estimators are based on high-dimension techniques including (a continuum of) ℓ1-penalized quantile regressions and low biased equations, which allows us to handle the potentially large number of variables. We build upon recent results to obtain valid choice of the penalty parameters and rates of convergence. These results are derived without any assumptions on the separation from zero and are uniformly valid across a wide-range of models. With the additional assumptions that the coefficients are well-separated from zero, we can consistently estimate the graph associated with the dependence structure by hard thresholding the proposed estimators. Further we show how QGM can be used to represent the tail interdependence of the variables which plays an important role in application concern with extreme events in opposition to average behavior. We show that the associated tail risk network can be used for measuring systemic risk contributions. We also apply the framework to study financial contagion and the impact of downside movement in the market on the dependence structure of assets’ return. Finally, we illustrate the properties of the proposed framework through simulated examples.Belloni, AlexandreWed, 07 Mar 2018 19:02:22 GMThttp://ageconsearch.umn.edu/record/2693212016Effects of agricultural pesticide utilization on farmers health in Egbeda Local Government Area, Oyo State, Nigeria
http://ageconsearch.umn.edu/record/268438
Occupational and environmental exposures to pesticides cause a wide range of human health problems. This study assessed the health effect of pesticide use in Egbeda Local Government Area of Oyo State with from socioeconomic point of view by identifying the various pesticides used, describing the health signs and symptoms associated with pesticide application and determining the health effects of pesticide application on farm households. The survey instrument was a well-structured questionnaire administered to respondents through multiple stage sampling technique. A total of 120 respondents were interviewed for the study. Data collected were analysed using descriptive statistics and Tobit regression model. The results showed that commonly used pesticides comprised of herbicides (61.67 %), fungicides (20.0%) and insecticides (18.33%). Majority (95%) reported that they or someone in their family had suffered from pesticide-related health signs and symptoms during or after application of pesticides. The effect of pesticides application on farmer’s health’s was captured by exposure, frequency of exposure and pesticide cocktails. The coefficient of exposure, frequency of exposure and pesticide cocktails were positive and significant at p<0.01, p<0.1and p<0.05 respectively. It is therefore recommended that the awareness of farmers and authorities needs to be raised regarding the use of protective equipment and correct procedures when handling pesticides and, also, that there should be stricter enforcement of existing pesticide regulation and monitoring policies to minimize the threats that farmers’ current practices pose to their health and the livelihood sustainability.Adekunle, C.P.Thu, 15 Feb 2018 19:21:48 GMThttp://ageconsearch.umn.edu/record/268438urn:ISSN:0794-47482017-10-01Nonparametric Seemingly Unrelated Regression
http://ageconsearch.umn.edu/record/267943
This papers presents a method for simultaneously estimating a system of nonparametric multiple regressions which may seem unrelated, but where the errors are potentially correlated between equations. We show that the prime advantage of estimating such a 'seemingly unrelated' system of nonparametric regressions is that substantially less observations can be required to obtain reliable function estimates than if each of the regression equations was estimated separately and the correlation ignored. This increase in efficiency is investigated empirically using both simulated and real data. The method suggested here develops a Bayesian hierarchical framework where the regression function is represented as a linear combination of a large number of basis terms, the number of which is typically greater than the sample size. All the regression coefficients, and the variance matrix of the errors, are estimated simultaneously using their posterior means. The computation is carried out using a Markov chain Monte Carlo sampling scheme that employs a 'focused sampling' step to combat the high dimensional representation of the function and a Metropolis-Hastings step to correctly account for the distribution of the covariance matrix. The methodology is also easily extended to other nonparametric multivariate regression models.Smith, MichaelTue, 06 Feb 2018 21:47:44 GMThttp://ageconsearch.umn.edu/record/2679431998Modified Likelihood and Related Methods for Handling Nuisance Parameters in the Linear Regression Model
http://ageconsearch.umn.edu/record/267941
In this paper, different approaches to dealing with nuisance parameters in likelihood based inference are presented and illustrated by reference to the linear regression model with nonspherical errors. The estimator of the error variance using each of the approaches is also derived for the linear regression model with spherical errors. We observe that many of these estimators are unbiased. A theoretical comparison of the likelihood functions is reported and we note that some of them are equivalent. Empirical evidence in the literature indicates that estimators based on the conditional profile likelihood and tests based on the marginal likelihood have better small sample properties compared to those based on other likelihood and message length functions.Laskar, Mizan R.Tue, 06 Feb 2018 21:40:06 GMThttp://ageconsearch.umn.edu/record/2679411998Estimating Long-Term Trends in Tropospheric Ozone Levels
http://ageconsearch.umn.edu/record/267938
This paper estimates the long-term trends in the daily maxima of tropospheric ozone at six sites around the state of Texas. The statistical methodology we use controls for the effects of meteorological variables because it is known that variables such as temperature, wind speed and humidity substantially affect the formation of tropospheric ozone. A nonparametric regression model is estimated in which a general trivariate surface is used to model the relationship between ozone and these meteorological variables because there is little, or no, theory to specify the functional dependence of ozone on these variables. The model also allows for the effects of wind direction and seasonality. Each function in the model is represented as a linear combination of basis functions located at all of the design points. A trivariate basis is used for the function representing the combined effect of temperature, wind speed and humidity, while univariate bases are used to represent the other functions in the model. To estimate the functions nonparametrically we use a Bayesian hierarchical framework with a fractional prior. Due to the high dimensional representation of the signal, a Markov chain Monte Carlo sampling scheme employing Gibbs sub-chains that 'focus' on the basis terms that are most likely to contribute to the signal is used to carry out the computations. We also estimate an appropriate data transformation simultaneously with the function estimates. The empirical results indicate that key meteorological variables explain most of the variation in daily ozone maxima through a nonlinear interaction and that their effects are consistent across the six sites. However, the estimated trends vary considerably from site to site, even within the same city. A simulation based on the design of the data indicates that the Bayesian approach is substantially more efficient than MARS (Friedman, 1991).Smith, MichaelTue, 06 Feb 2018 21:22:36 GMThttp://ageconsearch.umn.edu/record/2679381998Bayesian Semiparametric Regression: An Exposition and Application to Print Advertising Data
http://ageconsearch.umn.edu/record/267935
A new regression based approach is proposed for modeling marketing databases. The approach is Bayesian and provides a number of significant improvements over current methods. Independent variables can enter into the model in either a parametric or nonparametric manner, significant variables can be identified from a large number of potential regressors and an appropriate transformation of the dependent variable can be automatically selected from a discrete set of pre-specified candidate transformations. All these features are estimated simultaneously and automatically using a Bayesian hierarchical model coupled with a Gibbs sampling scheme. Being Bayesian, it is straightforward to introduce subjective information about the relative importance of each variable, or with regard to a suitable data transformation. The methodology is applied to print advertising Starch data collected from thirteen issues of an Australian women's monthly magazine. The empirical results highlight the complex and detailed relationships that can be uncovered using the methodology.Smith, MichaelTue, 06 Feb 2018 21:06:48 GMThttp://ageconsearch.umn.edu/record/2679351997Additive Nonparametric Regression with Autocorrelated Errors
http://ageconsearch.umn.edu/record/267921
A Bayesian approach is presented for nonparametric estimation of an additive regression model with autocorrelated errors. Each of the potentially nonlinear components is modelled as a regression spline using many knots, while the errors are modelled by a high order stationary autoregressive process parameterised in terms of its autocorrelations. The distribution of significant knots and partial autocorrelations is accounted for using subset selection. Our approach also allows the selection of a suitable transformation of the dependent variable. All aspects of the model are estimated simultaneously using Markov chain Monte Carlo. It is shown empirically that the proposed approach works well on a number of simulated and real examples.Smith, MichaelTue, 06 Feb 2018 19:49:11 GMThttp://ageconsearch.umn.edu/record/2679211996Comparisons of Estimators and Tests Based on Modified Likelihood and Message Length Functions
http://ageconsearch.umn.edu/record/267490
The presence of nuisance parameters causes unwanted complications in statistical and econometric inference procedures. A number of modified likelihood and message length functions have been developed for better handling of nuisance parameters but they are not equally efficient. In this paper, we empirically compare different modified likelihood and message length functions in the context of estimation and testing of parameters from linear regression disturbances that follow either first-order moving average or first-order autoregressive error processes. The results show that estimators based on the conditional profile likelihood and tests based on the marginal likelihood are best. If there is a minor identification problem, the sizes of the likelihood ratio and Wald tests based on simple message length functions are best. The true sizes of the Lagrange multiplier tests based on message length functions are rather poor because the score functions of message length functions are biased.Laskar, Mizan R.Wed, 31 Jan 2018 19:17:58 GMThttp://ageconsearch.umn.edu/record/2674901998