We propose and compare combinations of several methods for classifying transportation activity data from smartphone GPS and accelerometer sensors. We have two main objectives. First, we aim to classify our data as accurately as possible. Second, we aim to reduce the dimensionality of the data as much as possible in order to reduce the computational burden of the classification. We combine dimension reduction and classification algorithms and compare them with a metric that balances accuracy and dimensionality. In doing so, we develop a classification algorithm that accurately classifies five different modes of transportation (i.e., walking, biking, car, bus and rail) while being computationally simple enough to run on a typical smartphone. Further, we use data that required no behavioral changes from the smartphone users to collect. Our best classification model uses the random forest algorithm to achieve 96.8% accuracy.
Major League Baseball franchises expend an abundance of resources on scouting in preparation for the June Amateur Draft. In addition to the classic “tools” assessed, another factor considered is age: Younger players may get selected over older players of equal ability because of anticipated development. Additionally, Little League rules in effect until 2006 operated on an August 1–July 31 year meaning that, in their youth, players born on August 1 were the eldest in their cohort. We define a relative age to be how old a player was (in days) when the youngest player in their cohort was born. For example, a player born on July 31 would have a relative age of 1 day, whereas a player born on August 1 would have a relative age of 365 days. We examine the performance of high school (HS) players selected in the June Draft from 1987 to 2011. We find that more relatively old players are selected in the Draft. Conversely, both age and relative age have a significant negative relationship with the odds of reaching the major leagues. Given that a draftee reaches the majors, however, there is no difference based on age or relative age, as measured by games played, wins above replacement, and on-base plus slugging percentage. Had the draft market operated efficiently, neither relative age nor age on draft day would have captured additional variation in performance after controlling for draft position and other factors. We conclude that teams have undervalued both young and relatively young HS players.
1 March 2020 in Journal of Quantitative Analysis in Sports
We examine correlates of tenure length for professional soccer managers. Using 521 managers from Major League Soccer (MLS), Spain's La Liga, and the English Premier League (EPL) whose tenures occurred between 2000 and 2015, we assess the association between both performance-related and non-performance variables, and manager duration. Performance variables include measures of a team's ranking (or position) and relegation/promotion indicators. Non-performance variables include manager nationality and age, the timing of a manager's hire, and the team's wage bill. We employ survival analytic methods, including Cox's proportional hazards model, to explore the effects of fixed and time-dependent covariates on coach tenure length. We find that La Liga managers have shorter survival, as do managers who were older when they were hired. Furthermore, finishing with a better ranking and, more importantly, improving on previous team performance yields longer survival. Most strikingly, however, we find a significant disparity in the comparison of domestic and foreign managers within a league. While the difference in longevity between domestic and foreign managers in La Liga and the EPL was minimal, American managers in MLS survived significantly longer than their foreign peers.
Sporting careers observed over a preset time interval can be partitioned into two distinct subsamples. These samples consist of individuals whose careers had already commenced at the start of the time interval (prevalent subsample) and individuals whose careers began during the time interval (incident subsample) as well the respective individual-level covariate data such as salary, height, weight, performance statistics, draft position, etc. Under the assumption of a proportional hazards model, we propose a partial likelihood estimator to model the effect of covariates on survival via an adjusted risk set sampling procedure for when the incident cohort data is used in conjunction with the prevalent cohort data. We use simulated failure time data to validate the combined cohort proportional hazards methodology and illustrate our model using an NBA data set for career durations measured between 1990 and 2008.
7 January 2022 in BMC Medical Research Methodology
When modelling the survival distribution of a disease for which the symptomatic progression of the associated condition is insidious, it is not always clear how to measure the failure/censoring times from some true date of disease onset. In a prevalent cohort study with follow-up, one approach for removing any potential influence from the uncertainty in the measurement of the true onset dates is through the utilization of only the residual lifetimes. As the residual lifetimes are measured from a well-defined screening date (prevalence day) to failure/censoring, these observed time durations are essentially error free. Using residual lifetime data, the nonparametric maximum likelihood estimator (NPMLE) may be used to estimate the underlying survival function. However, the resulting estimator can yield exceptionally wide confidence intervals. Alternatively, while parametric maximum likelihood estimation can yield narrower confidence intervals, it may not be robust to model misspecification. Using only right-censored residual lifetime data, we propose a stacking procedure to overcome the non-robustness of model misspecification; our proposed estimator comprises a linear combination of individual nonparametric/parametric survival function estimators, with optimal stacking weights obtained by minimizing a Brier Score loss function.