JGJanusz Gajda

Janusz Gajda

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Janusz Gajda has a total of 37 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Effect of Global Energy Price Shocks on Dynamics of World Agricultural and Food Prices

    by Szczepan Figiel, Janusz Gajda, Justyna Kufel-Gajda

    Prices and quantities in agricultural commodity and food product markets are subject to constant changes due to evolving supply and demand conditions. Big and sudden shifts in supply or demand may lead to price movements that bring negative consequences for food producers or consumers. Factors causing such movements can be of different natures, but substantial changes in the world energy price levels are supposed to be one of the most important. The purpose of the study was to investigate the effect of global energy price shocks on the evolution of food commodities and food consumer prices. Using the World Bank data on the respective price indices, we looked for shocks in these data series by utilizing statistical tools. Having identified three global energy price shocks in the period 2000–2024 induced by the financial crisis of 2008, the COVID-19 pandemic, and the outbreak of war in Ukraine, their influence on the world agricultural commodity prices and food consumer prices was assessed. It was found that the series of energy, food commodity, and food consumer price indices were related in the long term. Also, the occurrence of global energy price shocks to a visible extent translated into global food commodity and food consumer price shocks. Applying various statistical and econometric techniques, including Chow tests and MS-VAR modelling, enables the identification of which breaking points led to regime changes between the analysed variables. The most sensitive to the structural breaking points appeared to be the relation between energy and consumer food prices. This discovery can be considered our major contribution.


  • Open Access

    Patent activity in the Polish food system and the business cycle. Conclusions from spectral analysis

    by Justyna Kufel-Gajda, Janusz Gajda

    Invention activity is an important innovation development factor, as it not only imposes changes in economic activity, but also often leads to the completely new structures within sectors. The aim of the paper was to investigate the relationship between macroeconomic and sectorial business cycle, and the patent filings in the food system in Poland in the period 2006–2022 taking advantage of statistics defined in time and frequency domains. Data on patent applications from 34 IPC classification (sub-)classes were analyzed, combined into five groups: food, beverages, tobacco, agricultural chemistry, agricultural machinery and packaging. The results obtained with spectral and cross-spectral analysis tools were compared with results obtained in the time domain. The positive relationship between patents and macroeconomic cycle is the strongest for cycles lasting 3.5–4 years, which dominate in reference to GDP series in the analyzed period. Patent filings are strongly positively correlated also with the agri-food sector output, especially in shorter cycles of 2.5–3 years. Although cycles of patent filings fluctuate much more than both business cycles, they seem to be more similar to the agri-food sector fluctuations. Moreover, patent activity may be treated as an early warning indicator of sectorial economic activity, whereas it’s lagging in reference to the aggregated economic activity.

  • Iterated tempered stable process

    by Ritik Soni, Janusz Gajda, Ashok Kumar Pathak

  • Open Access

    From Farms to Forests: An Exploration of Afforestation Efforts in Poland Under the Rural Development Programme (2007–2020)

    by Anna M. Klepacka, Andrzej A. Romaniuk, Janusz Gajda, Ewa E. Chećko

    Forests are important in mitigating climate change and addressing biodiversity loss. In Poland, where forest cover has steadily increased since World War II, afforestation of agricultural land has emerged as a key strategy supported by both EU and national policies. This study evaluates the implementation of Poland’s afforestation programs under the Rural Development Programme (RDP) for the periods of 2007–2013 and 2014–2020 using data provided by the Agency for Restructuring and Modernisation of Agriculture (ARMA) and focusing on the financial incentives offered to private landowners and regional variations in their uptake. Results show that afforestation under the RDP aligned with the EU climate neutrality goals, with a total of 37,721 hectares afforested and PLN 243.5 million handed as financial support, predominantly (81%) over the first studied period. The largest afforestation efforts were observed in the Mazowieckie and Warmińsko-Mazurskie voivodeships, with Warszawa, Poland’s capital, achieving the highest municipal afforestation area across both periods. Financial considerations influenced farmer participation, but rising land prices and historical land use patterns were contributing factors. Urban afforestation and increasing demand for woody biomass further highlight the benefits of afforestation. However, conifer-dominated stands, prevalent on low-quality land, can present challenges to ecosystem stability under future climate scenarios, necessitating diversification towards deciduous species. Additionally, low uptake during the 2014–2020 period showcases the need for improved incentives to bolster participation. Still, afforestation presents significant economic and environmental opportunities, advancing long-term policy objectives while addressing critical challenges in climate.


  • Distinguishing between fractional Brownian motion with random and constant Hurst exponent using sample autocovariance-based statistics

    by Aleksandra Grzesiek, Janusz Gajda, Samudrajit Thapa, Agnieszka Wyłomańska

    Fractional Brownian motion (FBM) is a canonical model for describing dynamics in various complex systems. It is characterized by the Hurst exponent, which is responsible for the correlation between FBM increments, its self-similarity property, and anomalous diffusion behavior. However, recent research indicates that the classical model may be insufficient in describing experimental observations when the anomalous diffusion exponent varies from trajectory to trajectory. As a result, modifications of the classical FBM have been considered in the literature, with a natural extension being the FBM with a random Hurst exponent. In this paper, we discuss the problem of distinguishing between two models: (i) FBM with the constant Hurst exponent and (ii) FBM with random Hurst exponent, by analyzing the probabilistic properties of statistics represented by the quadratic forms. These statistics have recently found application in Gaussian processes and have proven to serve as efficient tools for hypothesis testing. Here, we examine two statistics—the sample autocovariance function and the empirical anomaly measure—utilizing the correlation properties of the considered models. Based on these statistics, we introduce a testing procedure to differentiate between the two models. We present analytical and simulation results considering the two-point and beta distributions as exemplary distributions of the random Hurst exponent. Finally, to demonstrate the utility of the presented methodology, we analyze real-world datasets from the financial market and single particle tracking experiment in biological gels.


  • Open Access

    Non-Gaussian Measures in Infinite Dimensional Spaces: the Gamma-Grey Noise

    by Luisa Beghin, Lorenzo Cristofaro, Janusz Gajda

    In the context of non-Gaussian analysis, Schneider [29] introduced grey noise measures, built upon Mittag-Leffler functions; analogously, grey Brownian motion and its generalizations were constructed (see, for example, [6, 7, 9, 27]). In this paper, we construct and study a new non-Gaussian measure, by means of the incomplete-gamma function (exploiting its complete monotonicity). We label this measure Gamma-grey noise and we prove, for it, the existence of Appell system. The related generalized processes, in the infinite dimensional setting, are also defined and, through the use of the Riemann-Liouville fractional operators, the (possibly tempered) Gamma-grey Brownian motion is consequently introduced. A number of different characterizations of these processes are also provided, together with the integro-differential equation satisfied by their transition densities. They allow to model anomalous diffusions, mimicking the procedures of classical stochastic calculus.


  • Open Access

    Testing of two-dimensional Gaussian processes by sample cross-covariance function

    by Katarzyna Maraj-Zygmąt, Aleksandra Grzesiek, Grzegorz Sikora, Janusz Gajda, Agnieszka Wyłomańska

    In this paper, we address the issue of testing two-dimensional Gaussian processes with a defined cross-dependency structure. Multivariate Gaussian processes are widely used in various applications; therefore, it is essential to identify the theoretical model that accurately describes the data. While it is relatively straightforward to do so in a one-dimensional case, analyzing multi-dimensional vectors requires considering the dependency between the components, which can significantly affect the efficiency of statistical methods. The testing methodology presented in this paper is based on the sample cross-covariance function and can be considered a natural generalization of the approach recently proposed for testing one-dimensional Gaussian processes based on the sample autocovariance function. We verify the efficiency of this procedure on three classes of two-dimensional Gaussian processes: Brownian motion, fractional Brownian motion, and two-dimensional autoregressive discrete-time process. The simulation results clearly demonstrate the effectiveness of the testing methodology, even for small sample sizes. The theoretical and simulation results are supported by analyzing two-dimensional real-time series that describe the main risk factors of a mining company, namely, copper price and exchange rates (USDPLN). We believe that the introduced methodology is intuitive and relatively simple to implement, and thus, it can be applied in many real-world scenarios where multi-dimensional data are examined.


  • Open Access

    Multiclass Image Classification Using GANs and CNN Based on Holes Drilled in Laminated Chipboard

    by Grzegorz Wieczorek, Marcin Chlebus, Janusz Gajda, Katarzyna Chyrowicz, Kamila Kontna, Michał Korycki, Albina Jegorowa, Michał Kruk

    The multiclass prediction approach to the problem of recognizing the state of the drill by classifying images of drilled holes into three classes is presented. Expert judgement was made on the basis of the quality of the hole, by dividing the collected photographs into the classes: “very fine,” “acceptable,” and “unacceptable.” The aim of the research was to create a model capable of identifying different levels of quality of the holes, where the reduced quality would serve as a warning that the drill is about to wear down. This could reduce the damage caused by a blunt tool. To perform this task, real-world data were gathered, normalized, and scaled down, and additional instances were created with the use of data-augmentation techniques, a self-developed transformation, and with general adversarial networks. This approach also allowed us to achieve a slight rebalance of the dataset, by creating higher numbers of images belonging to the less-represented classes. The datasets generated were then fed into a series of convolutional neural networks, with different numbers of convolution layers used, modelled to carry out the multiclass prediction. The performance of the so-designed model was compared to predictions generated by Microsoft’s Custom Vision service, trained on the same data, which was treated as the benchmark. Several trained models obtained by adjusting the structure and hyperparameters of the model were able to provide better recognition of less-represented classes than the benchmark.