Jun Zhu

237D Russell Laboratories
1630 Linden Drive
Madison, WI 53706


PhD  Iowa State University, Ames, 2000 (Statistics)
MSE  Johns Hopkins University, Baltimore, 1995 (Mathematical Sciences)
BA  Knox College, Galesburg 1994 (Mathematics and Computer Science)

The main components of my research activities are statistical methodological research and scientific collaborative research.  My statistical methodological research concerns developing statistical methodology for analyzing spatially referenced data (spatial statistics) and spatial data repeatedly sampled over time (spatial-temporal statistics), that arise often in the physical, biological, and social sciences.   My collaborative research concerns applying modern statistical methods, especially spatial and spatial-temporal statistics, to studies of agricultural, biological, and ecological systems conducted by research scientists.  To a large extent, my overall research program involves a close connection between the two types of research activities: Problems in my collaborative research that do not have adequate statistical tools motivate my statistical methodological research; whereas the new methods I develop in statistical methodological research are applied in my collaborative projects. 


Statistics 571: Statistical Methods for Bioscience I

Statistics 572: Statistical Methods for Bioscience II

Statistics 575: Statistical Methods for Spatial Data

Statistics 992: Statistics for Spatial Data: Theory and Methods. 


Department of Statistics
Department of Entomology
Biometry Program, College of Agricultural and Life Sciences


Statistical Methodological Publications:
Cressie, N., Zhu, J., Baddeley, A.J., and Nair, M.G. (2000). Directed Markov point processes
as limits of partially ordered Markov models. Methodology and Computing in Applied
Probability, 2, 5–21.
Zhu, J., Lahiri, S.N., and Cressie, N. (2002). Asymptotic inference for spatial CDFs over time.
Statistica Sinica, 12, 843–861.
Zhu, J., Eickhoff, J.C., and Kaiser, M.S. (2003). Modeling the dependence between number of
trials and success probability in a beta-binomial–Poisson mixture distribution. Biometrics,
59, 957–963.
Eickhoff, J.C., Zhu, J., and Amemiya, Y. (2004). On the simulation size and the convergence of
the Monte Carlo EM algorithm via likelihood-based distances. Statistics and Probability
Letters, 67, 161–171.
Zhu, J., Morgan, C.L.S., Norman, J.M., Yue, W., and Lowery, B. (2004). Combined mapping
of soil properties using a multi-scale tree-structured spatial model. Geoderma, 118, 321–
Zhu, J. and Morgan, G.D. (2004). Comparison of spatial variables over subregions using a
block bootstrap. Journal of Agricultural, Biological, and Environmental Statistics, 9,
Zhu, J. and Morgan, G.D. (2004). A nonparametric procedure for analyzing repeated-measures
of spatially correlated data. Environmental and Ecological Statistics, 11, 431–443.
Tracey, J.A., Zhu, J., and Crooks, K. (2005). A set of nonlinear regression models for animal
movement in response to a single landscape feature. Journal of Agricultural, Biological,
and Environmental Statistics, 10, 1–18.
Zhu, J., Eickhoff, J.C., and Yan, P. (2005). Generalized linear latent variable models for
repeated measures of spatially correlated multivariate data. Biometrics, 61, 674–683.
Zhu, J., Huang, H.-C., and Wu, J. (2005). Modeling spatial-temporal binary data using
Markov random fields. Journal of Agricultural, Biological, and Environmental Statistics,
10, 212–225.
Zhu, J. and Yue, W. (2005). A multiresolution tree-structured spatial linear model. Journal
of Computational and Graphical Statistics, 14, 168–184.
Ives, A.R. and Zhu, J. (2006). Statistics for correlated data: phylogenies, space, and time.
Ecological Applications, 16, 20–32.
Lahiri, S.N. and Zhu, J. (2006). Resampling methods for spatial regression models under a
class of stochastic designs. Annals of Statistics, 34, 1774–1813.
Yue, W. and Zhu, J. (2006). On estimation and prediction for multivariate multiresolution
tree-structured models. Statistica Sinica, 16, 981–1020.
Zhu, J. and Lahiri, S.N. (2007). Bootstrapping the empirical distribution function of a spatial
process. Statistical Inference for Stochastic Processes, 10, 107–145.
Rasmussen, J.G., Møller, J., Aukema, B.H., Raffa, K.F., and Zhu, J. (2007). Bayesian inference
for multivariate point processes observed at sparsely distributed times. Journal of
the Royal Statistical Society Series B, 69, 701–713.
Chi, G. and Zhu, J. (2008). Spatial regression models for demographic analysis. Population
Research and Policy Review, 27, 17–42.
Zheng, Y. and Zhu, J. (2008). Markov chain Monte Carlo for a spatial-temporal autologistic
regression model. Journal of Computational and Graphical Statistics, 17, 123–137.
Zheng, Y., Zhu, J., and Li, D. (2008). Analyzing spatial panel data of cigarette demand: A
Bayesian hierarchical modeling approach. Journal of Data Science, 6, 467–489.
Zhu, J., Rasmussen, J.G., Møller, J., Aukema, B.H., and Raffa, K.F. (2008). Spatial-temporal
modeling of forest gaps generated by colonization from below- and above-ground bark
beetle species. Journal of the American Statistical Association, 103, 162–177.
Zhu, J., Zheng, Y., Carroll, A.L., and Aukema, B.H. (2008). Autologistic regression analysis
of spatial-temporal binary data via Monte Carlo maximum likelihood. Journal of
Agricultural, Biological, and Environmental Statistics, 13, 84–98.
Wang, H. and Zhu, J. (2009). Variable selection in spatial regression via penalized least
squares. Canadian Journal of Statistics, 37, 607–624.
Zheng, Y., Zhu, J., and Roy, A. (2010). Nonparametric Bayesian inference for the spectral
density function of a random field. Biometrika, 97, 238–245.
Zhu, J., Huang, H.-C., and Reyes, P.E. (2010). On selection of spatial linear models for lattice
data. Journal of the Royal Statistical Society Series B, 72, 389–402.
Tracey, J.A., Zhu, J., and Crooks, K. (2011). Modeling and inference of animal movement
using artificial neural network. Environmental and Ecological Statatistics. Accepted for
Lin, F. and Zhu, J. (2011). Additive hazards regression and partial likelihood estimation for
ecological monitoring data. Statistics and Its Inference. Accepted for publication.
Scientific Publications:
Morgan, G.D., MacGuidwin, A.E., Zhu, J., and Binning, L.K. (2002). Root lesion nematode
(Pratylenchus penetrans) population dynamics over a three-year crop rotation. Agronomy
Journal, 94, 1146–1155.
Morgan, G.D., Stevenson, W.R., MacGuidwin, A.E., Kelling, K.A., Binning, L.K., and Zhu,
J. (2002). Plant pathogen population dynamics in potato fields. Journal of Nematology,
34, 189–193.
Ingham, S.C., Vivio, L.L., Losinski, J.A., and Zhu, J. (2004). Manual shaking as an alternative
to mechanical stomaching in preparing ground meats for microbiological analysis. Food
Protection Trends, 24, 253–256.
Anderson, D.A., Turner, M.G., Forester, J.D., Zhu, J., Boyce, M.S., Beyer, H., and Stowell,
L. (2005). Scale-dependent summer habitat use for reintroduced elk (Cervus canadensis)
in Wisconsin, USA. Journal of Wildlife Management, 69, 298–310.
Ingham, S.C., Fanslau, M.A., Engel R., Breuer, J.R., Breuer, J.E., Wright, T.H., Reith-
Rozelle, J.K., and Zhu, J. (2005). Evaluation of fertilization-to-planting and fertilizationto-
harvest intervals for safe use of non-composted bovine manure in Wisconsin vegetable
production. Journal of Food Protection, 68, 1134–1142.
Ingham, S.C., Engel, R.A., Fanslau, M.A., Schoeller, E.L., Searls, G.A., Buege, D.R., and
Zhu, J. (2005). Fate of Staphylococcus aureus on vacuum-packaged ready-to-eat meat
products stored at 21oC. Journal of Food Protection, 68, 1911–1915.
Magle, S.B., Zhu, J., and Crooks, K. (2005). Behavioral responses to repeated human intrusion
by black-tailed prairie dogs (Cynomys Ludovicianus). Journal of Mammalogy, 86, 524–
Smithwick, E.A.H., Mack, M.C., Turner, M.G., Chapin III, F.S., Zhu, J., and Balser, T.C.
(2005). Spatial heterogeneity of ecosystem processes after severe fire in a black spruce
(Picea mariana) forest, Alaska. Biogeochemistry, 76, 517–537.
Aukema, B.H., Carroll, A.L., Zhu, J., Raffa, K.F., Sickley, T.A., and Taylor, S.W. (2006).
Landscape level analysis of mountain pine beetle in British Columbia, Canada: Spatiotemporal
development and spatial synchrony within the present outbreak. Ecography,
29, 427–441.
Algino, R.J., Ingham, S.C., and Zhu, J. (2007). Survey of antimicrobial effects of beef carcass
intervention treatments in very small state-inspected slaughter plants. Journal of Food
Science, 72, 173–179.
Ruby, J.R., Zhu, J., and Ingham, S.C. (2007). Using indicator bacteria and Salmonella spp.
test results from three large-scale beef abattoirs over an 18-month period to evaluate
intervention system efficacy and plan carcass testing for Salmonella spp. Journal of Food
Protection, 70, 2732–2740.
Vander-Zanden, M.J., Joppa, L.N., Allen, B.C., Chandra, S., Gilroy, D., Hogan, Z., Maxted,
J.T., and Zhu, J. (2007). Modeling spawning dates of Hucho taimen in Mongolia to
establish fishery management zones. Ecological Applications, 17, 2281–2289.
Aukema, B.H., Carroll, A.L., Zheng, Y., Zhu, J., Raffa, K.F., Moore, R.D., and Stahl, K.
(2008). Movement of outbreak populations of mountain pine beetle: Influence of spatiotemporal
patterns and climate. Ecography, 31, 348–358.
Mu˜noz, G.R., Kelling, K.A., Rylant, K.E., and Zhu, J. (2008) Field evaluation of nitrogen
availability from fresh and composted manure. Journal of Environmental Quality, 37,
Qin, X., Han, J., and Zhu, J. (2009). Spatial analysis of road weather safety data using a
Bayesian hierarchical modeling approach. Advances in Transportation Studies, 18, 69–
Aukema, B.H., Zhu, J., Møller, J., Rasmussen, J.G., and Raffa, K.F. (2010). Interactions
between below- and above-ground herbivores drive a forest decline and gap-forming syndrome.
Forest Ecology and Management, 259, 374–382.
Magle, S.B., Reyes, P., Zhu, J., and Crooks, K. (2010). Investigating local extinction, colonization, and habitat destruction for a keystone species in urban habitat. Biological
Conservation, 143, 2146-2155.

Book Review:
Zhu, J. (2006). Review of “Statistical Methods for Spatial Data Analysis” by O. Schabenberger
and C. A. Gotway. Journal of the American Statistical Association, 101, 389–390.


Resampling Methods

Earlier in my career, I worked on asymptotic inference for spatial cumulative distribution functions using spatial subsampling with application to environmental monitoring and assessment (Zhu et al. 2001, 2002).  I then expanded to statistical methodological research on spatial resampling methods.  On the one hand, I worked on the theory of the spatial block bootstrap, which is a resampling method alternative to spatial subsampling (Lahiri and Zhu 2006; Zhu and Lahiri 2007).  On the other hand, motivated by my collaborative research work with Dr. G. Morgan in Horticulture on the population dynamics of root lesion nematodes in potato fields (Morgan et al. 2002a, 2002b), I developed new statistical methodology based on the theory of the spatial block bootstrap. In particular, I developed a spatial block bootstrap for comparing spatial variables in different subregions (Zhu and Morgan 2004a) and for comparing spatial variables over time (Zhu and Morgan 2004b).

Multiresolution Models

From 2001 to 2006, I collaborated with Drs. J. Norman and C. Morgan in Soil Science on the development of the Precision Agricultural Landscape Modeling Systems (PALMS). I used a multiresolution spatial model for mapping soil properties in a computationally efficient manner (Zhu et al. 2004). Motivated by the PALMS work, Dr. W. Yue and I developed new statistical methodological research by extending the existing multiresolution spatial model to have a regression mean that allows for inclusion of covariates, which is more realistic and flexible, while maintaining computational efficiency using a change-of-resolution Kalman filter algorithm (Zhu et al. 2004; Zhu and Yue 2005; Yue and Zhu 2006).    

Animal Movement Models

Since 2001, I collaborated with Drs. K. Crooks, S. Magle, and  J. Tracey in Wildlife Ecology on the behavior and conservation of wildlife.  One research project concerns the conservation of mountain lions and other large mammals in southern California where the animals’ territories are threatened by rapid urban sprawl.   We quantified animal movement paths in terms of the angle and length of each move across landscape features over time by developing statistical nonlinear regression models and inference.  In the beginning, we considered the situation of one animal responding to one type of landscape feature and extended the traditional circular statistics to accommodate explanatory variables such as distance to a landscape feature (Tracey et al. 2005). We continued to extend the work to the situation of one animal responding to multiple types of landscape features, as well as from individual animal based inference to population-based inference (Tracey et al. 2011 forthcoming).  In another research project, we studied prairie dogs in the front range of Colorado. We compared prairie dogs in urban areas and those in rural areas using linear regression and logistic regression analysis (Magle et al. 2005) and assessed the impact of human activities on the colonization and extirpation of prairie dogs over time (Magle et al. 2010). 

Spatial-Temporal Statistics and Other Current Research

My current research interests are primarily in statistics for spatial-temporal data.  I have been collaborating with Drs. K. Raffa,  B. Aukema, and their research groups in Forest Entomology on population dynamics and interactions among trees, insects, and fungi in forest stands of Wisconsin.  The nature of the data is complex, involving spatial, temporal, and multiple response variables that are not necessarily normally distributed, which inspired me to pursue research in spatial-temporal statistics involving statistical methodology for spatial and temporal data  (Rasmussen et al. 2007; Zhu et al. 2008; Aukema et al. 2010).  During a visit to Academia Sinica in Taiwan in 2003, I worked with Drs. H.-C. Huang and J.-P. Wu on a spatial-temporal random field model for binary data with application to outbreaks of southern pine beetles (Zhu et al. 2005).  The modeling framework is an extension of the autologistic model for purely spatial data on a lattice.  Since then, my collaborators and I have been adapting the model for quantifying spatial-temporal patterns of mountain pine beetle outbreak in Western Canada (Aukema et al. 2006; Aukema et al. 2008; Zhu et al. 2008).  Furthermore, motivated by the fact that practical statistical tools are limited for complex spatial and spatial-temporal data that are not necessarily normal, I developed new statistical methodologies such as a latent variable model that has generalized linear models for the response variables and a spatial-temporal multivariate process for the latent variables (Zhu et al. 2005), continuous-time models (Rasmussen et al. 2007), and  general modeling frameworks for spatial-temporal binary data while addressing challenging computational issues (Zheng and Zhu 2008).

My current statistical methodology research focuses on model selection of spatial lattice models (Wang and Zhu 2009; Zhu et al. 2010) and time-to-event models for ecological monitoring (Rasmussen et al. 2007; Lin and Zhu 2011 forthcoming).   I also have continued interests in some of the more theoretical topics such as nonparametric Bayesian inference for spatial processes (Zheng et al. 2010) and asymptotic frameworks for maximum likelihood estimation in spatial models (Chu et al. 2011 in review; Zheng and Zhu 2011 in review). 

Applications in Ecological/Environmental Studies and Other Current Interests

Throughout the last ten years, I have been collaborating with research scientists in a wide variety of disciplines, especially in Wildlife Ecology, Conservation Biology, and Landscape Ecology.  Besides the aforementioned projects, I have also assisted the statistical modeling and analysis of the foraging behavior of green-backed herons (Zhu et al 2003), habitat selection by reintroduced elks in Northern Wisconsin (Anderson et al. 2005), effect of forest fires on ecosystems (Smithwick et al. 2005), spawning dates of Hucho taimen in Mongolia (Vander-Zanden et al. 2007), as well as general theory in quantitative ecology (Ives and Zhu 2006).  Most recently, I have expanded the areas of applications to Civil and Environmental Engineering such as spatial analysis of road weather safety data (Qin et al. 2009), Spatial Demography (Chi and Zhu 2008), and Environmental History (Jin et al. 2011 in review). 


The three courses I teach regularly are Statistics 571-Statistical Methods for Bioscience I, Statistics 572-Statistical Methods for Bioscience II, and Statistics 575-Statistical Methods for Spatial Data.  My objective in Stat571 is to provide research-oriented students in the agricultural, biological, and environmental sciences with a thorough grounding in the basic statistical methods.  My teaching philosophy is to stress an understanding of the procedures along with applications.  While keeping the mathematical complexities to a minimum, I give considerable attention to the analysis of real data.  I view the development of the ability to interpret results and to evaluate critically the methods used as of paramount importance.  My objective in Stat572 is to provide students in bioscience with a thorough understanding of modern statistical procedures.  Like in Stat571, I emphasize underlying concepts rather than an extensive coverage of a wide range of topics.  To a large extent the assignments involve the analysis of data sets that approach the real-world complexity of data encountered in research and substantial use is made of the computer in conducting such analyses.  The course Stat575 is directed towards graduate students who are interested in analyzing spatial data, including students from the environmental and ecological sciences, urban and regional planning, soil sciences, plant and animal sciences, and statistics.  Similar to Statistics 571 and 572, I focus mostly on statistical methods and stress an understanding of the underlying concepts, as opposed to simply providing a cookbook of statistical formulas.

I have designed and taught a seminar course Statistics 992-Statistics for Spatial Data: Theory and Methods.  The course is directed towards graduate students in statistics who are interested in learning statistics for analyzing spatial data.  The course is also suitable for students with strong quantitative skills in other disciplines such as atmospheric and oceanic sciences, earth sciences, ecology, economics, epidemiology, and mathematics.  Unlike Stat571, Stat572, and Stat575, I focus the course on not only spatial statistical methods, but also the underlying statistical theory.  Furthermore, I encourage critical and independent thinking of statistical concepts and help students improve communication skills.


Syndicate content