ivy's Picture

Ivy (I-Ming) Liu

BS (National Chengchi University), MS (Iowa State University), PhD (University of Florida),
Associate Professor
School of Mathematics and Statistics
Deputy Director
Centre for Data Science and Artificial Intelligence
Email: ivy.liu@vuw.ac.nz
Phone: +64 4 463 5648
ORCID: 0000-0002-3152-2632

Room: CO 424 Postal Address

Ivy Liu received a PhD in Statistics from University of Florida, Gainesville, Florida, USA. In 2000, she joined the Statistics and Operations Research group as a lecturer. Now, she is an associate professor in the School of Mathematics and Statistics at Victoria University of Wellington, New Zealand. Her main research area is in Categorical Data Analysis, including ordinal response data analysis, cluster analysis, longitudinal data analysis, repeated measurements, and their applications.


Research Interests

My main area of interest is categorical data analysis, particularly

Multidisciplinary Research Impact

  • Education and Psychometrics: Differential Item Functioning (DIF)

    DIF analysis is commonly used in large-scale educational tests and survey studies. DIF analysis helps reveal whether the test or survey is fair, or whether it is systematically biased against certain groups of test-takers, such as women or certain underprivileged groups. The Liu-Agresti estimator (Liu and Agresti, 1996, DOI: 10.2307/2532838) is widely used when performance is scored according to the degrees of proficiency, rather then correct or incorrect. The Liu-Agresti method can be implemented by packages DIFAS 5.0, WINSTEPS (DIF statistics), etc.

  • Econometrics and Biomedical Science: Blow-Up and Cluster (BUC)

    The BUC method (Mukherjee, Ahn, Liu, Rathouz, and Sanchez, 2008, DOI: 10.1002/sim.3325) is named by Baetschmann et al. (2015). For a longitudinal study, researchers are often concerned about the unobserved heterogeneity. The BUC method provides a simple way to obtain consistent fixed effect estimators for an ordered logit model to control the unobserved heterogeneity. The BUC method was originally proposed for biomedical case-control data. Gunasekara, Carter, Liu, , Richardson, and Blakely (2012) first used the method for a longitudinal data.

  • Social Science: Multiple response data

    A multiple response variable is a categorical variable for which respondents can select any number of outcome categories. Such data arise often in a survey. For example, a question might be ``which of the following learning methods have you used'' and the respondents could tick all items that apply. We (DOI: 10.1111/j.0006-341X.1999.00936.x, DOI: 10.1177/0049124101029004001, DOI: 10.1002/bimj.200710395, DOI: 10.1111/insr.12015) used maximum-likelihood or quasi-likelihood approaches to analyse such data. Some of the methods can be implemented using packages SPSS, R (MRCV), etc.

    Research Projects, Funding and Scholarships

    Contact me if you want to do a PhD/MSc in my research areas and want to get a scholarship.

  • Research Assistantships:

    Research publications

    R Codes


    In 2023, I am teaching the following courses: