Search for: All records

Many surveys collect information on discrete characteristics and continuous variables, that is, mixed-type variables. Small-area statistics of interest include means or proportions of the response variables as well as their domain means, which are the mean values at each level of a different categorical variable. However, item nonresponse in survey data increases the complexity of small-area estimation. To address this issue, we propose a multivariate mixed-effects model for mixed-type response variables subject to item nonresponse. We apply this method to two data structures where the data are missing completely at random by design. We use empirical data from two separate studies: a survey of pet owners and a dataset from the National Resources Inventory. In these applications, our proposed method leads to improvements relative to a direct estimator and a predictor based on a univariate model.

 

Estimates of resident satisfaction with public education have great utility in public administration, especially among decision makers in shrinking small communities. But such estimates are typically obtained via surveys, which are costly and often unreliable at high spatial resolutions given low response rates. Our study found that satisfaction with public schools among residents of small communities can be reasonably estimated at the community level using public data. Several models generalized adequately to unseen data—these models typically included the following covariates: state student assessment scores, school reorganizations, net open enrollment, and the cost of educational outcomes relative to neighboring districts. Our findings thus amount to a cost‐effective survey alternative for gauging satisfaction with public schools in small Iowa communities.

 

Abstract. Land surface temperature (LST) is one of the most important and widely used parameters for studying land surface processes. Moderate ResolutionImaging Spectroradiometer (MODIS) LST products (e.g., MOD11A1 and MYD11A1) can provide this information with moderate spatiotemporal resolution withglobal coverage. However, the applications of these data are hampered because of missing values caused by factors such as cloud contamination,indicating the necessity to produce a seamless global MODIS-like LST dataset, which is still not available. In this study, we used a spatiotemporalgap-filling framework to generate a seamless global 1 km daily (mid-daytime and mid-nighttime) MODIS-like LST dataset from 2003 to 2020based on standard MODIS LST products. The method includes two steps: (1) data pre-processing and (2) spatiotemporal fitting. In the datapre-processing, we filtered pixels with low data quality and filled gaps using the observed LST at another three time points of the same day. In thespatiotemporal fitting, first we fitted the temporal trend (overall mean) of observations based on the day of year (independent variable) in eachpixel using the smoothing spline function. Then we spatiotemporally interpolated residuals between observations and overall mean values for eachday. Finally, we estimated missing values of LST by adding the overall mean and interpolated residuals. The results show that the missing values inthe original MODIS LST were effectively and efficiently filled with reduced computational cost, and there is no obvious block effect caused by largeareas of missing values, especially near the boundary of tiles, which might exist in other seamless LST datasets. The cross-validation withdifferent missing rates at the global scale indicates that the gap-filled LST data have high accuracies with the average root mean squared error(RMSE) of 1.88 and 1.33∘, respectively, for mid-daytime (13:30) and mid-nighttime (01:30). The seamless global daily (mid-daytime andmid-nighttime) LST dataset at a 1 km spatial resolution is of great use in global studies of urban systems, climate research and modeling,and terrestrial ecosystem studies. The data are available at Iowa State University's DataShare at https://doi.org/10.25380/iastate.c.5078492 (T. Zhanget al., 2021). 

Abstract. Near-surface air temperature (Ta) is a key variable in global climatestudies. A global gridded dataset of daily maximum and minimum Ta (Tmax⁡ and Tmin⁡) is particularly valuable and critically needed inthe scientific and policy communities but is still not available. In this paper, we developed a global dataset of daily Tmax⁡ and Tmin⁡at 1 km resolution over land across 50∘ S–79∘ N from 2003 to 2020 through the combined use of ground-station-basedTa measurements and satellite observations (i.e., digital elevation model and land surface temperature) via a state-of-the-artstatistical method named Spatially Varying Coefficient Models with Sign Preservation (SVCM-SP). The root mean square errors in our estimates rangedfrom 1.20 to 2.44 ∘C for Tmax⁡ and 1.69 to 2.39 ∘C for Tmin⁡. We found that the accuracies were affectedprimarily by land cover types, elevation ranges, and climate backgrounds. Our dataset correctly represents a negative relationship betweenTa and elevation and a positive relationship between Ta and land surface temperature; it captured spatial and temporalpatterns of Ta realistically. This global 1 km gridded daily Tmax⁡ and Tmin⁡ dataset is the first of its kind, and weexpect it to be of great value to global studies such as the urban heat island phenomenon, hydrological modeling, and epidemic forecasting. The data havebeen published by Iowa State University at https://doi.org/10.25380/iastate.c.6005185 (Zhang and Zhou, 2022).