Skip to main content

Zeynep Işıl Kalaylıoğlu

 Prof. Zeynep Kalaylıoğlu

Room No: 235

Telephone: +90 (312) 210-5305

Fax No: +90 (312) 210-2959

E-mail: kzeynep@metu.edu.tr

CV || Personal Web Page || Git-Hub

 

 

Education:

  • Ph.D.: Department of Statistics, North Carolina State University, USA, 2002, thesis in Unit Root Testing in Stochastic Volatility Models (developed a method to test the volatility in financial series such as stock returns)
  • M.S.: Department of Statistics, North Carolina State University, USA, 1999, with minor degree in Computational Engineering
  • B.S.: Department of Statistics, Middle East Technical University, Türkiye, 1995, with minor courses from Computer Engineering

 

Academic and Professional Experience:

  • 2022 - Present Professor of Statistics
  • 2014 - 2022  Associate Professor of Statistics
  • 2009- 2014 Assistant Professor of Statistics, METU
  • 2007- 2009 Instructor, Department of Statistics, METU
  • 2002-2007 Cancer data scientist, IMS Inc. for the National Cancer Institute, USA

 

Administrative Experience:

  • 2025-present, Assistant to Head, Department of Statistics
  • 2024-present, Administrative Board Member, Ecosystem Implementation and Research Center METU
  • 2015-2019 Assistant to Head, Department of Statistics
  • 2014-2017 Assistant to Manager, Center for Wind Energy Researh, METU
  • 2012-2014 Assist. Prof. Representative Elect, Faculty Board, Faculty of Arts and Sciences

 

Research Focus:

My research focuses on developing Bayesian methods for applications in health and environmental sciences. I work on modeling complex spatially and temporally dynamic processes, aiming to better understand real-world phenomena through advanced statistical approaches. A key focus of my research is breast cancer, particularly investigating its risk factors and exploring their relationship with mammographic findings such as breast density. Through my work, I aim to contribute to improved risk assessment, deeper understanding of disease progression, and more informed decision-making in public health.

 

Teaching:

I  am experienced in teaching foundation of statistical theory, particularly mathematical statistics, statistical inference and theory of Bayesian analysis

 

Selected Publications:

  • Surucu, B. and Kalaylioglu, Z. (2026), A goodness-of-fit-test for GSSvM distribution (ongoing)
  • Danisman, F.B., Oflaz, Z. and Kalaylioglu, Z., Dogan, L., Kulturoglu, M. (2025), A latent Markov model  for mammogram density and breast cancer risk (in progress)
  • Kulturoglu, M.O, Aydin, F., Sagdic M.F., Aslan, F., Oflaz, Z., Danisman F.B., Kalaylioglu Z. and Dogan L. (2025), The impact of changes in breast density over time on breast cancer risk, Scientific Reports (Nature Portfolio), https://doi.org/10.1038/s41598-025-09315-1
    • Third most-cited journal in the world
    • In this study we developed a statistical approach to understand the relationship between routine mammography density data and breast cancer risk. Our study has shown that for women over 40, among those who had not yet reached Menopause, changes in breast density seen on Mammography—especially faster decreases—were linked to a higher risk of breast cancer, suggesting that tracking density changes might help tailor individual risk prevention.
  • Ciftci, S. and Kalaylioglu, Z. (2024), A Bayesian hybrid method for the analysis of generalized linear models with Missing-Not-At-Random Covariates, Journal of Statistical Computation and Simulation,
  • Kalaylioglu, Z. (2022), Analysis of correlated circular and extremal data with a novel flexible cylindrical distribution, Environmental and Ecological Statistics, 29, 207-222
    • In this study we developed a statistical model for analyzing circular-and-linear data (a type of Cylindrical Distribution) that helps researchers understand how things like wind speed and direction are linked, which can improve insights into real-world problems such as wildfire behavior. The model can be used to understand how environmental factors like highest wind speed, its direction, maximum temperature work together, helping improve computer simulations that predict how wildfires spread and how dangerous they might become.
  • Camli, O. , Kalaylioglu, Z., and SenGupta, A. (2022), Variable selection in linear-circular regression models, Journal of Applied Statistics, 50(16),  3337-3361
  • Kilic, M.B. , Kalaylioglu, Z., and SenGupta, A. (2022), A flexible Bayesian mixture approach for multi-modal circular data, HJMS,  1160-1173
  • Camli, O. and Kalaylioglu, Z. (2021), Bayesian predictive model selection in circular random effects models with applications in ecological and environmental studies, Environmental and Ecological Statistics, 28, 21-34
  • Hassanzadeh, F. and Kalaylioglu, Z. (2018), A new multimodal and asymmetric bivariate circular distribution, Environmental and Ecological Statistics, 25(3), 363-385
  • Tanju, O. and Kalaylioglu, Z. (2018), A cluster based model selection approach for logistic regression classifier, Journal of Statistical Computation and Simulation, 88(7), 1394-1414

    • In this paper, we developed a new method for choosing the best  Logistic Regression model that improves how well data can be classified compared to traditional approaches.

    • We then used it to identify women at higher or lower risk of breast cancer, helping support  for more informed healthcare decisions.

  • Kalaylioglu, Z. and Demirhan, H., (2017), A joint Bayesian approach for the analysis of response measured at a primary endpoint and longitudinal measurements, Statistical Methods in Medical Research, 26(6), 2885–2896
  • Karagulle, S. and Kalaylioglu, Z. (2016), A test for detecting etiologic heterogeneity in epidemiological studies, Journal of Applied Statistics, 43(3), 538-549
    • In this study, we developed a statistical testing method that can detect whether cancer risk factors affect different types of the disease in different ways, even when those differences are small.
    • We applied our method to breast cancer, the results show that smoking and breastfeeding duration influence different tumor subtypes differently—for example, smoking is more strongly linked to certain types of tumors, while breastfeeding duration relates more strongly to others.
  • Demirhan, H. and Kalaylioglu, Z. (2015), Joint prior distributions for variance parameters in normal hierarchical models, Journal of Multivariate Analysis, 135, 163-174

  • Kalaylioglu, Z. (2014), Performances of Bayesian model selection criteria for generalised linear models with nonignorably missing covariates, Journal of Statistical Computation and Simulation, 84(8),1670-1691

     

Certified trainings 

  • Hatha Yoga teacher (International Yoga Alliance)
  • Pranayama (Breathing) teacher (International Yoga Alliance)
  • Permaculture Designer (Australian Permaculture Institute)
  • Farming (olive and vineyard) (Türkiye Tarım Bakanlığı kayıtlı çiftçi)
  • Amateur Sailor (licensed both in USA and Türkiye)  and  wireless operator
  • First Aid (TC Sağlık Bakanlığı)
  • Patient and Elderly Care  (Sakarya Üniversitesi and Anadolu Üniversitesi)

Non-certified trainings

  • Permaculture (Oregon State University)
  • Whole Farm Sustainable Pest Management (Wild Farm Alliance)
  • Biodynamic Agriculture (Istafil Çiftliği)
  • Olive Oil tasting (Zeytindostu Derneği)
  • Landscape Architecture (Ankara Üniversitesi)
  • Latin dance (Washington DC)
  • Ashtanga yoga (Washington DC)
  • Flute and Acustic guitar (several private lessons in USA and Turkey)
  • Long time Club tennis player (Montgomery County Tennis Association, Sports International Bilkent, Ankara Tenis Kulübü, ODTÜ Tenis) (1994-2018)
  • Ongoing trainings in Tibetan Buddhism (FPMT centers)
Last updated