C88 - Data Collection and Data Estimation Methodology; Computer Programs: Other Computer SoftwareReturn

With the advancement of machine learning tools, an increasing number of algorithms are being utilized for predicting not only traditional time series data related to financial markets but also those connected to cryptocurrencies. This paper aims to compare various machine learning algorithms used for prediction, in order to identify the one with the greatest practical potential for creating a prediction model of Bitcoin’s price as an investment asset. The analysis focuses on supervised learning algorithms, taking into account the nature of the task involving long time series datasets. The paper also describes the exact process of creating and setting up individual models and their parameters, explaining procedures for obtaining and editing datasets, and shows how to evaluate performance of these models. In addition to the analysis of the main subject of research, which is Bitcoin, the paper also uses an analysis of reference cryptocurrencies such as Ethereum, Litecoin and NEO to compare the resulting performances. The processes consisting of editing the analysed datasets, creating individual prediction models, training and testing the performance of models on historical data, and creating, debugging and implementing individual machine learning models were realised through coding in the Python program.

In this article we analyze the continuous inflation dynamics using a four-equation model. When constructing the model, the traditional Kaldorian two-equation model is extended by adding two other equations. One of them describes an adaptive inflation expectations and the other continuous dynamics of the money market. In this setting, the instability velocity of money circulation is assumed due to the effects of expected inflation on money circulation velocity. Then the parameters of the model are estimated using the real Czech economic data. As it is a non-linear model in its parameters, a non-linear estimation technique is used for this purpose. Further, the stationarity as well as the stability of the estimated model is thoroughly examined as its instability may indicate that the model can generate some complex dynamics.

The paper deals with models and methods for evaluation of efficiency and performance of production units. Especially, it aims at data envelopment analysis (DEA) models, which are based on the definition of efficiency as the ratio of outputs produced by the unit on the one hand and inputs spent in the production process on the other hand. Standard data envelopment analysis models divide the units into inefficient and efficient ones but they make it not possible to classify the efficient units. That is why the paper deals with several definitions of super efficiency that are bases for classification of efficient units. The models are illustrated on a numerical example. The proposed approaches are applied in assessing the efficiency of branches of a Czech bank. Computer solution of DEA models and future research is discussed in conclusions.