Machine learning(ML)techniques have been widely used to address mental health questions.We discuss two main aspects of ML in psychiatry in this paper,that is,supervised learning and unsupervised learning.Examples are ...Machine learning(ML)techniques have been widely used to address mental health questions.We discuss two main aspects of ML in psychiatry in this paper,that is,supervised learning and unsupervised learning.Examples are used to illustrate how ML has been implemented in recent mental health research.展开更多
1.Introduction Survival analysis concerns the time from a well-defined origin to some event of interest, such as the time from surgery to death of a cancer patient, the time from wedding to divorce, and the time betwe...1.Introduction Survival analysis concerns the time from a well-defined origin to some event of interest, such as the time from surgery to death of a cancer patient, the time from wedding to divorce, and the time between the first and second suicide attempts.Although originated in research on the active lifetime of light bulbs and other electric devices, modern applications of survival analysis include many non-survival events.Thus, survival analysis may be more appropriately called the time-to-event analysis.展开更多
SUMMARY The p value has been widely used as a way to summarise the significance in data analysis. However, misuse and misinterpretation of the p value is common in practice. Our result shows that if the model specific...SUMMARY The p value has been widely used as a way to summarise the significance in data analysis. However, misuse and misinterpretation of the p value is common in practice. Our result shows that if the model specification is wrong, the distribution of the p value may be inappropriate, which makes the decision based on the p value invalid.展开更多
Diagnostic tests are usually based on some quantitative biomarkers.Two key parameters used to characterise the quality of a test are test sensitivity and specificity.Predictive values of the disease status based on te...Diagnostic tests are usually based on some quantitative biomarkers.Two key parameters used to characterise the quality of a test are test sensitivity and specificity.Predictive values of the disease status based on test results are also of interest in medical research and public health management.In this paper,we study the relations among sensitivity,specificity and predictive values of the test.The core concept is risk function,which is assumed to be an increasing function of the biomarker.Our results show that test sensitivity and specificity change in opposite directions.The positive predictive value and the sensitivity also change in opposite directions.Likewise,the negative predictive value and the specificity change in opposite directions.展开更多
Sample size justification is a very crucial part in the design of clinical trials. In this paper, the authors derive a new formula to calculate the sample size for a binary outcome given one of the three popular indic...Sample size justification is a very crucial part in the design of clinical trials. In this paper, the authors derive a new formula to calculate the sample size for a binary outcome given one of the three popular indices of risk difference.The sample size based on the absolute difference is the fundamental one, which can be easily used to derive sample size given the risk ratio or OR.展开更多
Suppose we have a sample of subjects in two treatment groups.To study the difference of the treatment effects,we can analyse the data using all subjects(overall analysis).We may also divide the subjects into several s...Suppose we have a sample of subjects in two treatment groups.To study the difference of the treatment effects,we can analyse the data using all subjects(overall analysis).We may also divide the subjects into several subgroups based on some covariates of interest(eg,gender),and study the treatment effects within each subgroup.The results of these two analyses may be different or even in opposite directions.In this paper,we give a general sufficient condition of consistency between the overall and subgroup analyses.展开更多
Mental health questions can be tackled through machine learning(ML)techniques.Apart from the two ML methods we introduced in our previous paper,we discuss two more advanced ML approaches in this paper:support vector m...Mental health questions can be tackled through machine learning(ML)techniques.Apart from the two ML methods we introduced in our previous paper,we discuss two more advanced ML approaches in this paper:support vector machines and artificial neural networks.To illustrate how these ML methods have been employed in mental health,recent research applications in psychiatry were reported.展开更多
Within the family of zero-inflated Poisson distributions, the data has Poisson distribution if any only if the mean equals the variance. In this paper we compare two closely related test statistics constructed based o...Within the family of zero-inflated Poisson distributions, the data has Poisson distribution if any only if the mean equals the variance. In this paper we compare two closely related test statistics constructed based on this idea. Our results show that although these two tests are asymptotically equivalent under the null hypothesis and are equally efficient, one test is always more efficient than the other one for small and medium sample sizes.展开更多
In this paper we study the relations of four possible generalized inverses of a general distribution functions and their right-continuity properties. We correct a right-continuity result of the generalized inverse use...In this paper we study the relations of four possible generalized inverses of a general distribution functions and their right-continuity properties. We correct a right-continuity result of the generalized inverse used in statistical literature. We also prove the validity of a new generalized inverse which is always right-continuous.展开更多
基金supported in part by the Novel Bio-statistical and Epidemiologic Methodology grants from the University of Rochester Medical Center Clinical and Translational Science Institute Pilot Awards Program
基金supported by a pilot grant(PI:Feng) from the Clinical and Translational Sciences Institute at the University of Rochester Medical Center(UR-CTSI GR500208)
文摘Machine learning(ML)techniques have been widely used to address mental health questions.We discuss two main aspects of ML in psychiatry in this paper,that is,supervised learning and unsupervised learning.Examples are used to illustrate how ML has been implemented in recent mental health research.
基金supported by the National Institute of Mental Health R01 MH075017-01A1by the National Institute of Mental Health R34 MH096854-02,both to Dr.Kerry L.Knox
文摘1.Introduction Survival analysis concerns the time from a well-defined origin to some event of interest, such as the time from surgery to death of a cancer patient, the time from wedding to divorce, and the time between the first and second suicide attempts.Although originated in research on the active lifetime of light bulbs and other electric devices, modern applications of survival analysis include many non-survival events.Thus, survival analysis may be more appropriately called the time-to-event analysis.
文摘SUMMARY The p value has been widely used as a way to summarise the significance in data analysis. However, misuse and misinterpretation of the p value is common in practice. Our result shows that if the model specification is wrong, the distribution of the p value may be inappropriate, which makes the decision based on the p value invalid.
文摘Diagnostic tests are usually based on some quantitative biomarkers.Two key parameters used to characterise the quality of a test are test sensitivity and specificity.Predictive values of the disease status based on test results are also of interest in medical research and public health management.In this paper,we study the relations among sensitivity,specificity and predictive values of the test.The core concept is risk function,which is assumed to be an increasing function of the biomarker.Our results show that test sensitivity and specificity change in opposite directions.The positive predictive value and the sensitivity also change in opposite directions.Likewise,the negative predictive value and the specificity change in opposite directions.
文摘Sample size justification is a very crucial part in the design of clinical trials. In this paper, the authors derive a new formula to calculate the sample size for a binary outcome given one of the three popular indices of risk difference.The sample size based on the absolute difference is the fundamental one, which can be easily used to derive sample size given the risk ratio or OR.
文摘Suppose we have a sample of subjects in two treatment groups.To study the difference of the treatment effects,we can analyse the data using all subjects(overall analysis).We may also divide the subjects into several subgroups based on some covariates of interest(eg,gender),and study the treatment effects within each subgroup.The results of these two analyses may be different or even in opposite directions.In this paper,we give a general sufficient condition of consistency between the overall and subgroup analyses.
文摘Mental health questions can be tackled through machine learning(ML)techniques.Apart from the two ML methods we introduced in our previous paper,we discuss two more advanced ML approaches in this paper:support vector machines and artificial neural networks.To illustrate how these ML methods have been employed in mental health,recent research applications in psychiatry were reported.
文摘Within the family of zero-inflated Poisson distributions, the data has Poisson distribution if any only if the mean equals the variance. In this paper we compare two closely related test statistics constructed based on this idea. Our results show that although these two tests are asymptotically equivalent under the null hypothesis and are equally efficient, one test is always more efficient than the other one for small and medium sample sizes.
文摘In this paper we study the relations of four possible generalized inverses of a general distribution functions and their right-continuity properties. We correct a right-continuity result of the generalized inverse used in statistical literature. We also prove the validity of a new generalized inverse which is always right-continuous.
