Journal of Data Science

We propose a Bayesian Negative Binomial-Bernoulli model to jointly analyze the patterns behind field goal attempts and the factors influencing shot success. We apply nonnegative CANDECOMP/PARAFAC tensor decomposition to study shot patterns and use logistic regression to predict successful shots. To maintain the conditional conjugacy of the model, we employ a double Pólya-Gamma data augmentation scheme and devise an efficient variational inference algorithm for estimation. The model is applied to shot chart data from the National Basketball Association, focusing on the regular seasons from 2015–16 to 2022–23. We consistently identify three latent features in shot patterns across all seasons and verify a popular claim from recent years about the increasing importance of three-point shots. Additionally, we find that the home court advantage in field goal accuracy disappears in the 2020–21 regular season, which was the only full season under strict COVID-19 crowd control, aside from the short bubble period in 2019–20. This finding contributes to the literature on the influence of crowd effects on home advantage in basketball games.

Abstract: The application of linear mixed models or generalized linear mixed models to large databases in which the level 2 units (hospitals) have a wide variety of characteristics is a problem frequently encountered in studies of medical quality. Accurate estimation of model parameters and standard errors requires accounting for the grouping of outcomes within hospitals. Including the hospitals as random effect in the model is a common method of doing so. However in a large, diverse population, the required assump tions are not satisfied, which can lead to inconsistent and biased parameter estimates. One solution is to use cluster analysis with clustering variables distinct from the model covariates to group the hospitals into smaller, more homogeneous groups. The analysis can then be carried out within these groups. We illustrate this analysis using an example of a study of hemoglobin A1c control among diabetic patients in a national database of United States Department of Veterans’ Affairs (VA) hospitals.

Abstract: PSA measurements are used to assess the risk for prostate cancer. PSA range and PSA kinetics such as PSA velocity have been correlated with in creased cancer detection and assist the clinician in deciding when prostate biopsy should be performed. Our aim is to evaluate the use of a novel, maxi mum likelihood estimation - prostate specific antigen (MLE-PSA) model for predicting the probability of prostate cancer using serial PSA measurements combined with PSA velocity in order to assess whether this reduces the need for prostate biopsy. A total of 1976 Caucasian patients were included. All these patients had at least 6 PSA serial measurements; all underwent trans-rectal biopsy with minimum 12 cores within the past 10 years. A multivariate logistic re gression model was developed using maximum likelihood estimation (MLE) based on the following parameters (age, at least 6 PSA serial measurements, baseline median natural logarithm of the PSA (ln(PSA)) and PSA velocity (ln(PSAV)), baseline process capability standard deviation of ln(PSA) and ln(PSAV), significant special causes of variation in ln(PSA) and ln(PSAV) detected using control chart logic, and the volatility of the ln(PSAV). We then compared prostate cancer probability using MLE-PSA to the results of prostate needle biopsy. The MLE-PSA model with a 50% cut-off probability has a sensitivity of 87%, specificity of 85%, positive predictive value (PPV) of 89%, and negative predictive value (NPV) of 82%. By contrast, a single PSA value with a 4ng/ml threshold has a sensitivity of 59%, specificity of 33%, PPV of 56%, and NPV of 36% using the same population of patients used to generate the MLE-PSA model. Based on serial PSA measurements, the use of the MLE-PSA model significantly (p-value < 0.0001) improves prostate cancer detection and reduces the need for prostate biopsy.

Abstract: In this study, the data based on nucleic acid amplification tech niques (Polymerase chain reaction) consisting of 23 different transcript vari ables which are involved to investigate genetic mechanism regulating chlamy dial infection disease by measuring two different outcomes of muring C. pneumonia lung infection (disease expressed as lung weight increase and C. pneumonia load in the lung), have been analyzed. A model with fewer reduced transcript variables of interests at early infection stage has been obtained by using some of the traditional (stepwise regression, partial least squares regression (PLS)) and modern variable selection methods (least ab solute shrinkage and selection operator (LASSO), forward stagewise regres sion and least angle regression (LARS)). Through these variable selection methods, the variables of interest are selected to investigate the genetic mechanisms that determine the outcomes of chlamydial lung infection. The transcript variables Tim3, GATA3, Lacf, Arg2 (X4, X5, X8 and X13) are being detected as the main variables of interest to study the C. pneumonia disease (lung weight increase) or C. pneumonia lung load outcomes. Models including these key variables may provide possible answers to the problem of molecular mechanisms of chlamydial pathogenesis.

Abstract: When comparing the performance of health care providers, it is important that the effect of such factors that have an unwanted effect on the performance indicator (eg. mortality) is ruled out. In register based studies randomization is out of question. We develop a risk adjustment model for hip fracture mortality in Finland by using logistic regression. The model is used to study the impact of the length of the register follow-up period on adjusting the performance indicator for a set of comorbidities. The comorbidities are congestive heart failure, cancer and diabetes. We also introduce an implementation of the minimum description length (MDL) principle for model selection in logistic regression. This is done by using the normalized maximum likelihood (NML) technique. The computational burden becomes too heavy to apply the usual NML criterion and therefore a technique based on the idea of sequentially normalized maximum likelihood (sNML) is introduced. The sNML criterion can be evaluated efficiently also for large models with large amounts of data. The results given by sNML are then compared to the corresponding results given by the traditional AIC and BIC model selection criteria. All three comorbidities have clearly an effect on hip fracture mortality. The results indicate that for congestive heart failure all available medical history should be used, while for cancer it is enough to use only records from half a year before the fracture. For diabetes the choice of time period is not as clear, but using records from three years before the fracture seems to be a reasonable choice.