Journal of Data Science

Land use land cover (LULC) change in the agriculture, is a critical area of concern as it directly impacts food security, environmental health, and economic stability. One of the leading LULC data products is the U.S. Department of Agriculture’s (USDA) Cropland Data Layer (CDL). Produced annually by the USDA National Agricultural Statistics Service (NASS) using satellite imagery, the CDL provides crop-specific data with an estimated classification accuracy of 85% to 95% for major crop types across the U.S. However, several limitations inherent to the CDL, such as crop underestimation bias, pixel misclassification, and difficulty distinguishing certain vegetation types, have raised questions about the accuracy of LULC change estimates derived from this dataset. In this paper, we introduce the R package cdlsim, designed to quantify the sensitivity of CDL-derived metrics through simulations of CDL data at the patch level using NASS published accuracy statistics. We present a case study utilizing landscape metrics calculated with the popular landscapemetrics R package to demonstrate the utility of cdlsim in quantifying the sensitivity of metrics to random perturbations in the data. The case study examines a mixed agricultural and grassland landscape in South Dakota, illustrating how our package enables researchers to achieve a more nuanced representation of land-use change.

Abstract: Simulation studies are important statistical tools used to inves-tigate the performance, properties and adequacy of statistical models. The simulation of right censored time-to-event data involves the generation of two independent survival distributions, where the rst distribution repre-sents the uncensored survival times and the second distribution represents the censoring mechanism. In this brief report we discuss how we can make it so that the percentage of censored data is previously de ned. The described method was used to generate data from a Weibull distribution, but it can be adapted to any other lifetime distribution. We further presented an R code function for generating random samples, considering the proposed approach.

Abstract: In any sport competition, there is a strong interest in knowing which team shall be the champion at the end of the championship. Besides this, the end result of a match, the chance of a team to be qualified for a specific tournament, the chance of being relegated, the best attack, the best defense, among others, are also subject of interest. In this paper we present a simple method with good predictive quality, easy implementation, low computational effort, which allows the calculation of all the interesting quantities above. Following Lee (1997), we estimate the average goals scored by each team by assuming that the number of goals scored by a team in a match follows a univariate Poisson distribution but we consider linear models that express the sum and the difference of goals scored in terms of five covariates: the goal average in a match, the home-team advantage, the team’s offensive power, the opponent team’s defensive power and a crisis indicator. The methodology is applied to the 2008-2009 English Premier League.

Overdispersion is a common phenomenon in Poisson modelling. The generalized Poisson (GP) distribution accommodates both overdispersion and under dispersion in count data. In this paper, we briefly overview different overdispersed and zero-inflated regression models. To study the impact of fitting inaccurate model to data simulated from some other model, we simulate data from ZIGP distribution and fit Poisson, Generalized Poisson (GP), Zero-inflated Poisson (ZIP), Zero-inflated Generalized Poisson (ZIGP) and Zero-inflated Negative Binomial (ZINB) model. We compare the performance of the estimates of Poisson, GP, ZIP, ZIGP and ZINB through mean square error, bias and standard error when the samples are generated from ZIGP distribution. We propose estimators of parameters of ZIGP distribution based on the first two sample moments and proportion of zeros referred to as MOZE estimator and compare its performance with maximum likelihood estimate (MLE) through a simulation study. It is observed that MOZE are almost equal or even more efficient than that of MLE of the parameters of ZIGP distribution.