Censored regression models have received a great deal of attention in both the theoretical and ap... more Censored regression models have received a great deal of attention in both the theoretical and applied econometric literature. Most of the existing estimation procedures for either cross sectional or panel data models are designed only for models with xed censoring. In this paper, a new procedure for adapting these estimators designed for xed censoring to models with random censorship is proposed. This procedure is then applied to the CLAD and quantile estimators of Powell(1984Powell( ,1986a to obtain an estimator of the regression coe cients under a mild conditional quantile restriction on the error term that is applicable to samples exhibiting xed or random censoring. The resulting estimator is shown to have desirable asymptotic properties, and performs well in a small scale simulation study. JEL Classi cation: C24, C14, C13. . We are grateful to Jambe and Rodchenko for helpful comments on an earlier draft.
Censored regression models have received a great deal of attention in both the theoretical and ap... more Censored regression models have received a great deal of attention in both the theoretical and applied econometric literature. Most of the existing estimation procedures for either cross sectional or panel data models are designed only for models with xed censoring. In this paper, a new procedure for adapting these estimators designed for xed censoring to models with random censorship is proposed. This procedure is then applied to the CLAD and quantile estimators of Powell(1984Powell( ,1986a to obtain an estimator of the regression coe cients under a mild conditional quantile restriction on the error term that is applicable to samples exhibiting xed or random censoring. The resulting estimator is shown to have desirable asymptotic properties, and performs well in a small scale simulation study. JEL Classi cation: C24, C14, C13. . We are grateful to Jambe and Rodchenko for helpful comments on an earlier draft.
Aim  The aim of this study was to systematically review the literature on the influence of enhanc... more Aim  The aim of this study was to systematically review the literature on the influence of enhanced recovery after surgery (ERAS) protocols and health related quality of life (HQoL) and patient satisfaction.Method  A systematic review of the literature from January 1990 to February 2009 was undertaken. Studies were included if they compared HQoL and/or patient satisfaction after ERAS and conventional surgery (CS). Jadad score was used to evaluate the studies. Results were divided into immediate (first week), early (second to third week) and late (more than 30 days after surgery) post-operative phases. A meta-analysis was not possible due to the heterogeneity of the studies.Results  Ten publications were included in the final analysis. In the first week after surgery, two non-randomised trials found reduced fatigue and another 2 non-randomised studies found reduced pain with ERAS. One randomised study found increased emotional distress on SF36 in ERAS patients. At two to three weeks after surgery, none of the multidimensional HQoL measures showed any differences. Increuscd fatigue was reported with CS in 2 studies. Limitations in activities of daily living were more marked after CS in one study. Beyond 30 days after surgery, none of the HQoL measures showed any differences. Only one non-randomised study compared patient satisfaction after ERAS and CS and no difference was found.Conclusion  There is no evidence that ERAS adversely affect HQoL or patient satisfaction. Certain aspects of HQoL such as pain and fatigue may improve with ERAS. Further research is required, especially in the early post-operative period.
Background  Enhanced recovery after surgery (ERAS) protocols aim to reduce postoperative lengths ... more Background  Enhanced recovery after surgery (ERAS) protocols aim to reduce postoperative lengths of stay (LOS). Laparoscopic surgery shares the same objective. Whether or not laparoscopic surgery conveys additional benefits over and above those achievable with ERAS alone is unclear.Method  A review of the literature from January 1966 to June 2008 was performed. Randomized trial and case series investigating the role of laparoscopic surgery within ERAS protocols were included. These were assessed for LOS, readmissions, return of gut function, quality of life and cost of treatment.Results  Four publications from three randomized trials and another seven case series were identified. Two randomized trials revealed no difference in the LOS and readmission rates between ERAS when used alone and ERAS with the inclusion of laparoscopic surgery. However, a third trial revealed a significantly shorter LOS and a lower readmission rate in the ERAS and laparoscopic group.Conclusion  The limited evidence presently available suggests that no further benefit is obtained by the inclusion of laparoscopic surgery in ERAS protocols. Further research is needed to derive more definitive conclusions.
Refeeding syndrome (RFS) describes the biochemical changes, clinical manifestations, and complica... more Refeeding syndrome (RFS) describes the biochemical changes, clinical manifestations, and complications that can occur as a consequence of feeding a malnourished catabolic individual. RFS has been recognised in the literature for over fifty years and can result in serious harm and death. Crude estimates of incidence, morbidity, and mortality are available for specific populations. RFS can occur in any individual but more commonly occurs in at-risk populations. Increased awareness amongst healthcare professionals is likely to reduce morbidity and mortality. This review examines the physiology of RFS and describes the clinical manifestations. A management strategy is described. The importance of a multidisciplinary approach is emphasized.
... 1 I would like to thank Shahrukh Khan for his help in conceptualizing the paper. I would like... more ... 1 I would like to thank Shahrukh Khan for his help in conceptualizing the paper. I would like to thank Solava Ibrahim and Luba Fakhrutdinova for their able research assistants. Milford Bateman and Philipp Lepenies made helpful comments on the first draft of the paper. ...
... 1 I would like to thank Shahrukh Khan for his help in conceptualizing the paper. I would like... more ... 1 I would like to thank Shahrukh Khan for his help in conceptualizing the paper. I would like to thank Solava Ibrahim and Luba Fakhrutdinova for their able research assistants. Milford Bateman and Philipp Lepenies made helpful comments on the first draft of the paper. ...
Journal of The American Statistical Association, 2005
... Gordon B. Dahl is Associate Professor of Economics (E-mail: [email protected]) andSh... more ... Gordon B. Dahl is Associate Professor of Economics (E-mail: [email protected]) andShakeeb Khan is Associate Professor of ... Lewbel and Linton (2002) esti-mated a nonparametric censored regression model with a fixed censoring point based on a mean restriction on ...
In this paper, we consider estimation of discrete response models exhibiting conditional heterosk... more In this paper, we consider estimation of discrete response models exhibiting conditional heteroskedasticity of a multiplicative form, where the latent error term is assumed to be the product of an unknown scale function and a homoskedastic error term. It is ÿrst shown that for estimation of the slope coe cients in a binary choice model under this type of restriction, the semiparametric information bound is zero, even when the homoskedastic error term is parametrically speciÿed. Hence, it is impossible to attain the parametric convergence rate for the parameters of interest. However, for ordered response models where the response variable can take at least three di erent values, the parameters of interest can be estimated at the parametric rate under the multiplicative heteroskedasticity assumption. Two estimation procedures are proposed. The ÿrst estimator, based on a parametric restriction on the homoskedastic component of the error term, is a two-step maximum likelihood estimators, where the unknown scale function is estimated nonparametrically in the ÿrst stage. The second procedure, which does not require the parametric restriction, estimates the parameters by a kernel weighted least-squares procedure. Under regularity conditions which are standard in the literature, both estimators are shown to be √ n-consistent and asymptotically normal.
It has been shown by Powell (1986a,b) that p n-consistent estimation of the slope parameters in t... more It has been shown by Powell (1986a,b) that p n-consistent estimation of the slope parameters in the linear censored regression model is possible under a conditional quantile and a conditional symmetry restriction on the error term, respectively. While the proposed estimators have desirable asymptotic properties, simulation studies have shown these estimators to exhibit a small sample bias in the opposite
Censored Least Absolute Deviations (CLAD) estimator for the censored linear regression model has ... more Censored Least Absolute Deviations (CLAD) estimator for the censored linear regression model has been regarded as a desirable alternative to maximum likelihood estimation methods due to its robustness to conditional heteroskedasticity and distributional misspeci cation of the error term. However, the CLAD estimation procedure has failed in certain empirical applications due to the restrictive nature of the \full rank" condition it requires. This condition can be especially problematic when the data is heavily censored. In this paper we introduce estimation procedures for heteroskedastic censored linear regression models with a much weaker identi cation restriction than that required for the CLAD, and which are exible enough to allow for various degrees of censoring. The new estimators are shown to have desirable asymptotic properties and perform well in small scale simulation studies, and can thus be considered as viable alternatives for estimating censored regression models, especially for applications in which the CLAD fails. JEL Classi cation: C14,C23,C24 . We are grateful to B.E. Honor e, J.L. Powell and two anonymous referees for their helpful comments.
This paper considers estimation of a sample selection model subject to conditional heteroskedasti... more This paper considers estimation of a sample selection model subject to conditional heteroskedasticityin both the selection and outcome equations. The form of heteroskedasticity allowed for ineach equation is multiplicative, and each of the two scale functions is left unspecified, except formild regularity conditions such as smoothness and boundedness. A three step estimator for theparameters of interest in the outcome equation
... Hong Kong University of Science and Technology Shakeeb Khan¤ University of Rochester ... One ... more ... Hong Kong University of Science and Technology Shakeeb Khan¤ University of Rochester ... One of the disadvantages of working with nonparametric conditional quantile estimators as opposed to nonparametric conditional mean estimators, is that the former is not in the ...
Censored Least Absolute Deviations (CLAD) estimator for the censored linear regression model has ... more Censored Least Absolute Deviations (CLAD) estimator for the censored linear regression model has been regarded as a desirable alternative to maximum likelihood estimation methods due to its robustness to conditional heteroskedasticity and distributional misspeci cation of the error term. However, the CLAD estimation procedure has failed in certain empirical applications due to the restrictive nature of the \full rank" condition it requires. This condition can be especially problematic when the data is heavily censored. In this paper we introduce estimation procedures for heteroskedastic censored linear regression models with a much weaker identi cation restriction than that required for the CLAD, and which are exible enough to allow for various degrees of censoring. The new estimators are shown to have desirable asymptotic properties and perform well in small scale simulation studies, and can thus be considered as viable alternatives for estimating censored regression models, especially for applications in which the CLAD fails. JEL Classi cation: C14,C23,C24 . We are grateful to B.E. Honor e, J.L. Powell and two anonymous referees for their helpful comments.
We develop semiparametric and instrumental-variables approaches to self-selection problems in com... more We develop semiparametric and instrumental-variables approaches to self-selection problems in comparisons of wage inequality between educational groups. We propose new estimators to identify the causal effect of college education on scale parameters of wage distributions, using symmetry conditions on the joint distribution of outcome and selection errors, along with kernel weighting schemes. A simulation study indicates that the proposed estimators perform well in finite samples. We illustrate these methods with a well-known study by , using college proximity as an instrument for schooling. Estimation results suggest that college education significantly increased the degree of wage inequality in 1976.
In this paper we propose an estimation procedure for a censored regression model where the latent... more In this paper we propose an estimation procedure for a censored regression model where the latent regression function has a partially linear form+ Based on a conditional quantile restriction, we estimate the model by a two stage procedure+ The first stage nonparametrically estimates the conditional quantile function at insample and appropriate out-of-sample points, and the second stage involves a simple weighted least squares procedure+ The proposed procedure is shown to have desirable asymptotic properties under regularity conditions that are standard in the literature+ A small scale simulation study indicates that the estimator performs well in moderately sized samples+
Censored regression models have received a great deal of attention in both the theoretical and ap... more Censored regression models have received a great deal of attention in both the theoretical and applied econometric literature. Most of the existing estimation procedures for either cross sectional or panel data models are designed only for models with xed censoring. In this paper, a new procedure for adapting these estimators designed for xed censoring to models with random censorship is proposed. This procedure is then applied to the CLAD and quantile estimators of Powell(1984Powell( ,1986a to obtain an estimator of the regression coe cients under a mild conditional quantile restriction on the error term that is applicable to samples exhibiting xed or random censoring. The resulting estimator is shown to have desirable asymptotic properties, and performs well in a small scale simulation study. JEL Classi cation: C24, C14, C13. . We are grateful to Jambe and Rodchenko for helpful comments on an earlier draft.
Censored regression models have received a great deal of attention in both the theoretical and ap... more Censored regression models have received a great deal of attention in both the theoretical and applied econometric literature. Most of the existing estimation procedures for either cross sectional or panel data models are designed only for models with xed censoring. In this paper, a new procedure for adapting these estimators designed for xed censoring to models with random censorship is proposed. This procedure is then applied to the CLAD and quantile estimators of Powell(1984Powell( ,1986a to obtain an estimator of the regression coe cients under a mild conditional quantile restriction on the error term that is applicable to samples exhibiting xed or random censoring. The resulting estimator is shown to have desirable asymptotic properties, and performs well in a small scale simulation study. JEL Classi cation: C24, C14, C13. . We are grateful to Jambe and Rodchenko for helpful comments on an earlier draft.
Aim  The aim of this study was to systematically review the literature on the influence of enhanc... more Aim  The aim of this study was to systematically review the literature on the influence of enhanced recovery after surgery (ERAS) protocols and health related quality of life (HQoL) and patient satisfaction.Method  A systematic review of the literature from January 1990 to February 2009 was undertaken. Studies were included if they compared HQoL and/or patient satisfaction after ERAS and conventional surgery (CS). Jadad score was used to evaluate the studies. Results were divided into immediate (first week), early (second to third week) and late (more than 30 days after surgery) post-operative phases. A meta-analysis was not possible due to the heterogeneity of the studies.Results  Ten publications were included in the final analysis. In the first week after surgery, two non-randomised trials found reduced fatigue and another 2 non-randomised studies found reduced pain with ERAS. One randomised study found increased emotional distress on SF36 in ERAS patients. At two to three weeks after surgery, none of the multidimensional HQoL measures showed any differences. Increuscd fatigue was reported with CS in 2 studies. Limitations in activities of daily living were more marked after CS in one study. Beyond 30 days after surgery, none of the HQoL measures showed any differences. Only one non-randomised study compared patient satisfaction after ERAS and CS and no difference was found.Conclusion  There is no evidence that ERAS adversely affect HQoL or patient satisfaction. Certain aspects of HQoL such as pain and fatigue may improve with ERAS. Further research is required, especially in the early post-operative period.
Background  Enhanced recovery after surgery (ERAS) protocols aim to reduce postoperative lengths ... more Background  Enhanced recovery after surgery (ERAS) protocols aim to reduce postoperative lengths of stay (LOS). Laparoscopic surgery shares the same objective. Whether or not laparoscopic surgery conveys additional benefits over and above those achievable with ERAS alone is unclear.Method  A review of the literature from January 1966 to June 2008 was performed. Randomized trial and case series investigating the role of laparoscopic surgery within ERAS protocols were included. These were assessed for LOS, readmissions, return of gut function, quality of life and cost of treatment.Results  Four publications from three randomized trials and another seven case series were identified. Two randomized trials revealed no difference in the LOS and readmission rates between ERAS when used alone and ERAS with the inclusion of laparoscopic surgery. However, a third trial revealed a significantly shorter LOS and a lower readmission rate in the ERAS and laparoscopic group.Conclusion  The limited evidence presently available suggests that no further benefit is obtained by the inclusion of laparoscopic surgery in ERAS protocols. Further research is needed to derive more definitive conclusions.
Refeeding syndrome (RFS) describes the biochemical changes, clinical manifestations, and complica... more Refeeding syndrome (RFS) describes the biochemical changes, clinical manifestations, and complications that can occur as a consequence of feeding a malnourished catabolic individual. RFS has been recognised in the literature for over fifty years and can result in serious harm and death. Crude estimates of incidence, morbidity, and mortality are available for specific populations. RFS can occur in any individual but more commonly occurs in at-risk populations. Increased awareness amongst healthcare professionals is likely to reduce morbidity and mortality. This review examines the physiology of RFS and describes the clinical manifestations. A management strategy is described. The importance of a multidisciplinary approach is emphasized.
... 1 I would like to thank Shahrukh Khan for his help in conceptualizing the paper. I would like... more ... 1 I would like to thank Shahrukh Khan for his help in conceptualizing the paper. I would like to thank Solava Ibrahim and Luba Fakhrutdinova for their able research assistants. Milford Bateman and Philipp Lepenies made helpful comments on the first draft of the paper. ...
... 1 I would like to thank Shahrukh Khan for his help in conceptualizing the paper. I would like... more ... 1 I would like to thank Shahrukh Khan for his help in conceptualizing the paper. I would like to thank Solava Ibrahim and Luba Fakhrutdinova for their able research assistants. Milford Bateman and Philipp Lepenies made helpful comments on the first draft of the paper. ...
Journal of The American Statistical Association, 2005
... Gordon B. Dahl is Associate Professor of Economics (E-mail: [email protected]) andSh... more ... Gordon B. Dahl is Associate Professor of Economics (E-mail: [email protected]) andShakeeb Khan is Associate Professor of ... Lewbel and Linton (2002) esti-mated a nonparametric censored regression model with a fixed censoring point based on a mean restriction on ...
In this paper, we consider estimation of discrete response models exhibiting conditional heterosk... more In this paper, we consider estimation of discrete response models exhibiting conditional heteroskedasticity of a multiplicative form, where the latent error term is assumed to be the product of an unknown scale function and a homoskedastic error term. It is ÿrst shown that for estimation of the slope coe cients in a binary choice model under this type of restriction, the semiparametric information bound is zero, even when the homoskedastic error term is parametrically speciÿed. Hence, it is impossible to attain the parametric convergence rate for the parameters of interest. However, for ordered response models where the response variable can take at least three di erent values, the parameters of interest can be estimated at the parametric rate under the multiplicative heteroskedasticity assumption. Two estimation procedures are proposed. The ÿrst estimator, based on a parametric restriction on the homoskedastic component of the error term, is a two-step maximum likelihood estimators, where the unknown scale function is estimated nonparametrically in the ÿrst stage. The second procedure, which does not require the parametric restriction, estimates the parameters by a kernel weighted least-squares procedure. Under regularity conditions which are standard in the literature, both estimators are shown to be √ n-consistent and asymptotically normal.
It has been shown by Powell (1986a,b) that p n-consistent estimation of the slope parameters in t... more It has been shown by Powell (1986a,b) that p n-consistent estimation of the slope parameters in the linear censored regression model is possible under a conditional quantile and a conditional symmetry restriction on the error term, respectively. While the proposed estimators have desirable asymptotic properties, simulation studies have shown these estimators to exhibit a small sample bias in the opposite
Censored Least Absolute Deviations (CLAD) estimator for the censored linear regression model has ... more Censored Least Absolute Deviations (CLAD) estimator for the censored linear regression model has been regarded as a desirable alternative to maximum likelihood estimation methods due to its robustness to conditional heteroskedasticity and distributional misspeci cation of the error term. However, the CLAD estimation procedure has failed in certain empirical applications due to the restrictive nature of the \full rank" condition it requires. This condition can be especially problematic when the data is heavily censored. In this paper we introduce estimation procedures for heteroskedastic censored linear regression models with a much weaker identi cation restriction than that required for the CLAD, and which are exible enough to allow for various degrees of censoring. The new estimators are shown to have desirable asymptotic properties and perform well in small scale simulation studies, and can thus be considered as viable alternatives for estimating censored regression models, especially for applications in which the CLAD fails. JEL Classi cation: C14,C23,C24 . We are grateful to B.E. Honor e, J.L. Powell and two anonymous referees for their helpful comments.
This paper considers estimation of a sample selection model subject to conditional heteroskedasti... more This paper considers estimation of a sample selection model subject to conditional heteroskedasticityin both the selection and outcome equations. The form of heteroskedasticity allowed for ineach equation is multiplicative, and each of the two scale functions is left unspecified, except formild regularity conditions such as smoothness and boundedness. A three step estimator for theparameters of interest in the outcome equation
... Hong Kong University of Science and Technology Shakeeb Khan¤ University of Rochester ... One ... more ... Hong Kong University of Science and Technology Shakeeb Khan¤ University of Rochester ... One of the disadvantages of working with nonparametric conditional quantile estimators as opposed to nonparametric conditional mean estimators, is that the former is not in the ...
Censored Least Absolute Deviations (CLAD) estimator for the censored linear regression model has ... more Censored Least Absolute Deviations (CLAD) estimator for the censored linear regression model has been regarded as a desirable alternative to maximum likelihood estimation methods due to its robustness to conditional heteroskedasticity and distributional misspeci cation of the error term. However, the CLAD estimation procedure has failed in certain empirical applications due to the restrictive nature of the \full rank" condition it requires. This condition can be especially problematic when the data is heavily censored. In this paper we introduce estimation procedures for heteroskedastic censored linear regression models with a much weaker identi cation restriction than that required for the CLAD, and which are exible enough to allow for various degrees of censoring. The new estimators are shown to have desirable asymptotic properties and perform well in small scale simulation studies, and can thus be considered as viable alternatives for estimating censored regression models, especially for applications in which the CLAD fails. JEL Classi cation: C14,C23,C24 . We are grateful to B.E. Honor e, J.L. Powell and two anonymous referees for their helpful comments.
We develop semiparametric and instrumental-variables approaches to self-selection problems in com... more We develop semiparametric and instrumental-variables approaches to self-selection problems in comparisons of wage inequality between educational groups. We propose new estimators to identify the causal effect of college education on scale parameters of wage distributions, using symmetry conditions on the joint distribution of outcome and selection errors, along with kernel weighting schemes. A simulation study indicates that the proposed estimators perform well in finite samples. We illustrate these methods with a well-known study by , using college proximity as an instrument for schooling. Estimation results suggest that college education significantly increased the degree of wage inequality in 1976.
In this paper we propose an estimation procedure for a censored regression model where the latent... more In this paper we propose an estimation procedure for a censored regression model where the latent regression function has a partially linear form+ Based on a conditional quantile restriction, we estimate the model by a two stage procedure+ The first stage nonparametrically estimates the conditional quantile function at insample and appropriate out-of-sample points, and the second stage involves a simple weighted least squares procedure+ The proposed procedure is shown to have desirable asymptotic properties under regularity conditions that are standard in the literature+ A small scale simulation study indicates that the estimator performs well in moderately sized samples+
Uploads
Papers by shakeeb khan