Presentation of my R package mergm, which can be used for social network data analysis.

I gave a talk about my extensions of the ERGM, where I also presented my algorithm.

My poster presentation in Warsaw was recognized with the best poster award.

Robust regression is an alternative to least squares regression when data are “polluted” with outliers or influential observations, and it can also be used to detect influential observations.

Social Network Data Analysis (SNDA) is an emerging field in statistics in the recent years, and the Exponential Random Graph Model (ERGM) is a promising approach for modelling network data...

Logistic regression also called a logit model, is used to model dichotomous outcome variables. In the logit model, the log odds of the outcome are modeled as a linear combination of the predictor variables.

We address this issue using a Generalised Additive Model (GAM) developed on humerus measurements of unborn and very young sheep of known age housed in modern collections.

We examine the inclusion of specific nodal random effects for first- and second-mode nodes for bipartite networks. The predicted nodal random effects deliver reasonable interpretation to understand the network behavior.

We introduce a new estimation framework in ERGMs iterating between pseudolikelihood and maximum likelihood estimation approaches for nodal effects and network statistics.

This explorative study, applies a new methodological approach and considers "carry or not" as a decision process.

The milk yield of the patients (cows) was documented on several days, and the rumen contraction was determined on two days.

Sentiment analysis determines the mood in social media related to products, services, campaigns, and companies. If opinions are predominantly negative, the company can analyze the reasons and respond...

Classification is a method of supervised learning where it is sufficient to derive a so-called class from the data, where the prediction is a class, i.e., categorical data is sought.

Prediction task is a method of supervised learning where we want to predict the target feature precisely as possible given some data.

The projects presented on this site rely on resources from external providers to correctly display data, e.g. mathematical formulas. As these external providers might access your personal information -e.g. IP addresses- when providing their resources, we need your permission to load these resources from external servers. Find out more about external resources used on this site on our data privacy page.