Search

High-Order Markov Chains (HOMC) are conventional models, based on transition probabilities, that are used by the United States Department of Agriculture (USDA) National Agricultural Statistics Service (NASS) to study crop-rotation patterns over time. However, HOMCs routinely suffer from sparsity and identifiability issues because the categorical data are represented as indicator (or dummy) variables. In fact, the dimension of the parametric space increases exponentially with the order of HOMCs required for analysis. While parsimonious representations reduce the number of parameters, as has been shown in the literature, they often result in less accurate predictions. Most parsimonious models are trained on big data structures, which can be compressed and efficiently processed using alternative algorithms. Consequently, a thorough evaluation and comparison of the prediction results obtain using a new HOMC algorithm and different types of Deep Neural Networks (DNN) across a range of agricultural conditions is warranted to determine which model is most appropriate for operational crop specific land cover prediction of United States (US) agriculture. In this paper, six neural network models are applied to crop rotation data between 2011 and 2021 from six agriculturally intensive counties, which reflect the range of major crops grown and a variety of crop rotation patterns in the Midwest and southern US. The six counties include: Renville, North Dakota; Perkins, Nebraska; Hale, Texas; Livingston, Illinois; McLean, Illinois; and Shelby, Ohio. Results show the DNN models achieve higher overall prediction accuracy for all counties in 2021. The proposed DNN models allow for the ingestion of long time series data, and robustly achieve higher accuracy values than a new HOMC algorithm considered for predicting crop specific land cover in the US.

The emerging sectors of agriculture, such as organics, urban, and local food, tend to be dominated by farms that are smaller, more transient, more diverse, and more dispersed than the traditional farms in the rural areas of the United States. As a consequence, a list frame of all farms within one of these sectors is difficult to construct and, even with the best of efforts, is incomplete. The United States Department of Agriculture’s (USDA’s) National Agricultural Statistics Service (NASS) maintains a list frame of all known and potential U.S. farms and uses this list frame as the sampling frame for most of its surveys. Traditionally, NASS has used its area frame to assess undercoverage. However, getting a good measure of the incompleteness of the NASS list frame using an area frame is cost prohibitive for farms in these emerging sectors that tend to be located within and near urban areas. In 2016, NASS conducted the Local Food Marketing Practices (LFMP) survey. Independent samples were drawn from (1) the NASS list frame and (2) a web-scraped list of local food farms. Using these two samples and capture–recapture methods, the total number and sales of local food operations at the United States, regional, and state levels were estimated. To our knowledge, the LFMP survey is the first survey in which a web-scraped list frame has been used to assess undercoverage in a capture–recapture setting to produce official statistics. In this article, the methods are presented, and the challenges encountered are reviewed. Best practices and open research questions for conducting surveys using web-scraped list frames and capture–recapture methods are discussed.

The United States Department of Agriculture's National Agricultural Statistics Service (NASS) uses probability surveys of hog owners to estimate quarterly hog inventories in the United States at the national and state levels. NASS also receives data from external sources. A panel of commodity experts forms the Agricultural Statistics Board (ASB). The ASB establishes the NASS official estimates for each quarter by taking into account survey estimates and other relevant sources of information that are available in numerical and non‐numerical form. The aim of this article is to propose an estimation method of hog inventories by combining the NASS proprietary survey results, the hog transaction data, the past ASB panel expert analyses, biological dynamics, and the inter‐inventory relationship constraints. This approach downscales the official estimates to provide monthly estimates according to well‐defined biological growth patterns. The model developed in this study provides national estimates that may inform the quarterly reports.

When conducting a national survey or census, administrative data may be available that can provide reliable values for some of the variables. Survey and census estimates should be consistent with reliable administrative data. Calibration can be used to improve the estimates by further adjusting the survey weights so that estimates of targeted variables honor bounds obtained from administrative data. The commonly used methods of calibration produce non-integer weights. For the Census of Agriculture, estimates of farms are provided as integers so as to insure consistent estimates at all aggregation levels; thus, the calibrated weights are rounded to integers. The calibration and rounding procedure used for the 2012 Census of Agricultural produced final weights that were substantially different from the survey weights that had been adjusted for under-coverage, non-response, and misclassification. A new method that calibrates and rounds as a single process is provided. The new method produces integer, calibrated weights that tend to be consistent with more calibration targets and are more correlated with the modeled census weights. In addition, the new method is more computationally efficient. Supplementary materials accompanying this paper appear online.

The reconstruction of hydro-stratigraphic units in subsoil (a general term indicating all the materials below ground level) plays an important role in the assessment of soil heterogeneity, which is a keystone in groundwater flow and transport modeling. A geostatistical approach appears to be a good way to reconstruct subsoil, and now other methods besides the classical indicator (co)kriging are available as alternative approximations of the conditional probabilities. Some of these techniques take specifically into account categorical variables as lithologies, but they are computationally prohibitive. Moreover, the stage before subsoil prediction/simulation can be very informative from a hydro-stratigraphic point of view, as the detailed transiogram analysis of this paper demonstrates. In this context, an application of the spMC package for the R software is presented by using a test site located within the Venetian alluvial plain (NE Italy). First, a detailed transiogram analysis was conducted, and then a maximum entropy approach, based on transition probabilities, named Markovian-type Categorical Prediction (MCP), was applied to approximate the posterior conditional probabilities. The study highlights some advantages of the presented approach in term of hydrogeological knowledge and computational efficiency. The spMC package couples transiogram analysis with a maximum entropy approach by taking advantage of High-Performance Computing (HPC) techniques. These characteristics make the spMC package useful for simulating hydro-stratigraphic units in subsoil, despite the use of a large number of lithologies (categories). The results obtained by spMC package suggest that this software should be considered a good candidate for simulating subsoil lithological distributions, especially of limited areas.

Heart failure is a major cause for premature death. Given the heterogeneity of the heart failure syndrome, identifying genetic determinants of cardiac function and structure may provide greater insights into heart failure. Despite progress in understanding the genetic basis of heart failure through genome wide association studies, the heritability of heart failure is not well understood. Gaining further insights into mechanisms that contribute to heart failure requires systematic approaches that go beyond single trait analysis. We integrated a Bayesian multi-trait approach and a Bayesian networks for the analysis of 10 correlated traits of cardiac structure and function measured across 3387 individuals with whole exome sequence data. While using single-trait based approaches did not find any significant genetic variant, applying the integrative Bayesian multi-trait approach, we identified 3 novel variants located in genes, RGS3, CHD3, and MRPL38 with significant impact on the cardiac traits such as left ventricular volume index, parasternal long axis interventricular septum thickness, and mean left ventricular wall thickness. Among these, the rare variant NC_000009.11:g.116346115C > A (rs144636307) in RGS3 showed pleiotropic effect on left ventricular mass index, left ventricular volume index and maximal left atrial anterior-posterior diameter while RGS3 can inhibit TGF-beta signaling associated with left ventricle dilation and systolic dysfunction.

The paper presents the spatial Markov Chains (spMC) R-package and a case study of subsoil simulation/prediction located in a plain site of Northeastern Italy. spMC is a quite complete collection of advanced methods for data inspection, besides spMC implements Markov Chain models to estimate experimental transition probabilities of categorical lithological data. Furthermore, simulation methods based on most known prediction methods (as indicator Kriging and CoKriging) were implemented in spMC package. Moreover, other more advanced methods are available for simulations, e.g. path methods and Bayesian procedures, that exploit the maximum entropy. Since the spMC package was developed for intensive geostatistical computations, part of the code is implemented for parallel computations via the OpenMP constructs. A final analysis of this computational efficiency compares the simulation/prediction algorithms by using different numbers of CPU cores, and considering the example data set of the case study included in the package.