Forskning
Multivariate analysis stands as a powerful methodological framework in statistical research, enabling investigators to examine complex datasets characterized by multiple interdependent variables simultaneously. This analytical approach has become indispensable across diverse scientific disciplines where understanding intricate relationships between numerous factors is essential for drawing valid conclusions. Its applications span numerous fields including econometrics, psychometrics, environmental science, marketing and consumer research, genomics and bioinformatics, and social sciences, demonstrating its versatility in addressing complex research questions.
The rapid advancement of the Internet of Things (IoT) has precipitated an exponential growth in data availability, creating both significant challenges and novel opportunities for statistical methodology. This data revolution has particularly highlighted the need for advanced high-dimensional analysis techniques, as traditional methods often prove inadequate when dealing with modern datasets characterized by large numbers of variables relative to observations. This evolving landscape necessitates the development of innovative statistical approaches tailored to address the unique demands of high-dimensional data analysis.
My research program focuses on two interconnected strands: methodological development for high-dimensional statistical analysis and applied data science in energy-related domains. On the methodological front, I develop robust statistical techniques specifically designed for high-dimensional settings, where conventional approaches may fail. Complementing this theoretical work, I investigate practical applications of machine learning and data science methodologies to address pressing challenges in the energy sector, with particular emphasis on developing solutions that bridge the gap between advanced analytics and industrial implementation.
Vi är kvalitetscertifierade
Ekonomihögskolan och Linnéuniversitetet är ackrediterade av The Association to Advance Collegiate Schools of Business, AACSB.
Mina forskargrupper
-
Deterministic and Stochastic Modelling Forskningsområdet Deterministic and Stochastic Modelling inom Linnaeus University Centre for Data Intensive Sciences and Applications (DISA) samlar forskare med…
-
Linnaeus University Centre for Data Intensive Sciences and Applications DISA är Linnéuniversitetets spetsforskningsmiljö som arbetar med insamling, analys och nyttogörande av stora datamängder.…
Mina pågående forskningsprojekt
Publikationer
Artikel i tidskrift (Refereegranskat)
-
Lundgren, M., Hough, R.L., Beesley, L., Troldborg, M., Trakal, L., et al. (2025). Modeling metal uptake by selected vegetables from urban soils in Europe : uncovering key soil factors using partial least squares regression (PLS-R). Human and Ecological Risk Assessment. 31 (3-4). 434-458.
Status: Publicerad -
Dai, D., Javed, F., Karlsson, P.S., Månsson, K. (2025). Nonlinear forecasting with many predictors using mixed data sampling kernel ridge regression models. Annals of Operations Research.
Status: Epub för tryck -
Dai, D., Hao, C., Jin, S., Liang, Y. (2025). Regularized estimation of Kronecker structured covariance matrix using modified Cholesky decomposition. Journal of Statistical Computation and Simulation. 95 (5). 905-930.
Status: Publicerad -
Dai, D., Pan, J., Liang, Y. (2022). Regularized estimation of the Mahalanobis distance based on modified Cholesky decomposition. Communications in Statistics: Case Studies, Data Analysis and Applications. 8 (4). 559-573.
Status: Publicerad -
Dai, D., Liang, Y. (2021). High-Dimensional Mahalanobis Distances of Complex Random Vectors. Mathematics. 9 (16).
Status: Publicerad -
Yan, J., Karlsson, A., Zou, Z., Dai, D., Edlund, U. (2020). Contamination of heavy metals and metalloids in biomass and waste fuels : Comparative characterisation and trend estimation. Science of the Total Environment. 700. 1-19.
Status: Publicerad -
Dai, D. (2020). Mahalanobis distances on factor model based estimation. Econometrics. 8 (1). 1-11.
Status: Publicerad -
Dai, D., Holgersson, T., Karlsson, P.S. (2017). Expected and unexpected values of Individual Mahalanobis Distances. Communications in Statistics - Theory and Methods. 46 (18). 8999-9006.
Status: Publicerad -
Gregebo, B., Dai, D., Schillberg, B., Baehr, M., Nyström, B., et al. (2014). Private and non-private disc herniation patients : do they differ?. The Open Orthopaedics Journal. 8. 237-241.
Status: Publicerad
Kapitel i bok, del av antologi (Refereegranskat)
- Liang, Y., Hao, C., Dai, D. (2024). Two-sample intraclass correlation coefficient tests for matrix-valued data. Statistical Modeling and Applications : Heavy-Tailed, Skewed Distributions and Mixture Modeling, Volume 2. Springer.
- Liang, Y., Dai, D. (2020). On Explicit Estimation of the Growth Curve Model with a Block Circular Covariance Structure. Recent Developments in Multivariate and Random Matrix Analysis : Festschrift in Honour of Dietrich von Rosen. Springer. 255-266.
Konferensbidrag (Refereegranskat)
- Dai, D., Holgersson, T. (2018). High-Dimensional CLTs for Individual Mahalanobis Distances. Trends and perspectives in linear statistical inference : proceedings of the LINSTAT2016 meeting held 22-25 August 2016 in Istanbul, Turkey. 57-68.