Search results

The COVID-19 pandemic deteriorated the economic situation and raised the issue of the quality of banks’ assets and, in particular, the growth of non-performing loans (NPLs). The study approaches a topical subject that is of interest to banks and society at large, as credit availability is likely to be reduced. Over the last 10 years, the Baltic countries’ banking sector has significantly improved its risk management policies and practices, increased capital ratios on its balance sheets, and created risk reserves. The current chapter examines the factors affecting NPLs in the Baltic States based on advanced econometric modelling applied to data extracted from the International Monetary Fund (IMF) and Eurostat. The study results show that credit risk management in the Baltic States has significantly improved compared to the period before the global financial crisis (GFC), the capitalisation of credit institutions is one of the highest in the European Union (EU), and banks are liquid and profitable. Lending recovered from the downturn in the first phase of the pandemic, and credit institutions have taken advantage of the European Central Bank’s (ECB) long-term funding programme ITRMO III to improve the liquidity outlook. Although the credit quality of commercial banks has not deteriorated, as the exposures of credit institutions in the most affected sectors are insignificant and governments have provided fiscal support to businesses and households, some challenges remain. The increase in credit risk is expected due to rising production prices as well as the rebuilding of disrupted supply chains. The findings allow conclusions to be drawn on the necessary actions to mitigate the credit risk of the banking sector.

The purpose of the article is to study the recent tendencies of growth of Russia’s agro-industrial complex (AIC), determine the optimal scenario of its development, and develop recommendations in the sphere of state regulation for its practical implementation. While there are tendencies of growing production and increase in Russia’s export, against this background, there is a tendency of quicker increase of import of food – if it continues, positive balance of foreign trade of food products in 2018 will turn into negative balance in 2020–2024. Though efficiency of crop farming is peculiar for a tendency of quick growth, efficiency of animal breeding is stable, which does not allow overcoming the growing deficit of food in Russia, which grows under the influence of the tendency of wear of fixed funds and slow implementation of new fixed funds due to insufficient financing. Scenarios of mid-term (i.e., until 2024) growth of Russia’s AIC are compiled, of which the most optimal is scenario that requires technological advancements, due to which increase in the value of index of food security up to 85.00 points (27%) will be achieved and the set goals of growth and development of Russia’s AIC will be reached. For a successful optimal scenario of the growth of Russia’s AIC, we offer recommendations in the sphere of state regulation of its digital modernization: adoption of the national strategy of transition to AIC 4.0 within the program “Digital economy of the RF,” development of import substitution in the AIC with emphasis on B2B markets, preparation of the technological platform for transition to AIC 4.0, and sufficient financing for digital modernization of the AIC.

Introduction: The insurance industry has unprecedented growth, and the demand for insurance has outgrown in the recent past due to the prevailing pandemic. The companies have a large base of the data set at their disposal, and companies must appropriately handle these data to come out with valuable solutions. Data mining enables insurance companies to gain an insightful approach to map strategies and gain competitive advantage, thus strengthening the profits that will allow them to identify the effectiveness of back-propagation neural network (BPNN) and support vector machines (SVMs) for the companies considered under study. Data mining techniques are the data-driven extraction techniques of information from large data repositories, thus discovering useful patterns from the voluminous data (Weiss & Indurkya, 1998).

Purpose: The present study is performed to investigate the comparative performance of BPNNs and SVMs for the selected Indian insurance companies.

Methodology: The study is conducted by extracting daily data of Indian insurance companies listed on the CNX 500. The data were then transformed into technical indicators for predictive model building using BPNN and SVMs. The daily data of the selected insurance companies for four years, that is, 1 April 2017 to 21 March 2021, were used for this. The data were further transformed into 90 data sets for different periods by categorising them into biannual, annual, and two-year collective data sets. Additionally, the comparison was made for the models generated with the help of BPNNs and SVMs for the six Indian insurance companies selected under this study.

Findings: The findings of the study exhibited that the predictive performance of the BPNN and SVM models are significantly different from each other for SBI data, General Insurance Corporation of India (GICRE) data, HDFC data, New India Assurance Company Ltd. (NIACL) data, and ICICIPRULI data at a 5% level of significance.

Introduction: The insurance industry is one of the lucrative sectors of the economy. However, it is volatile because of the large chunk of data generated by the transactions taking place daily. However, every bit of it is responsible for creating market trends for stock investors to predict the returns. The specialised data mining techniques act as a solution for decision-making, reducing uncertainty in decision-making.

Purpose: There are limited studies that have examined the efficiency and effectiveness of data mining techniques across the companies in the insurance industry to date. To enable the companies to take exact benefit of data mining techniques in insurance, the present study will focus on investigating the efficiency of artificial neural network (ANN) and support vector machine SVM across insurance companies of CNX 500.

Method: For predictive models, various technical indicators were considered independent variables, and change in return, i.e. increase and decrease, was deemed a dependent variable. The indicators were transformed from daily raw data of insurance company’s stock values spanning four years. We formed 90 data sets of varied periods for building the model – specifically six months, one year, two years, and four years for selected six insurance companies.

Findings: The study’s findings revealed that ANN performed best for the ICICIPRULI data model in terms of hit ratio. Whereas the performance of SVM was observed to be the best for the ICICIGI data model. In the case of pairwise comparison among the six selected Indian insurance companies from CNX 500, the extracted data evaluated and concluded that there were eight significantly different pairs based on hit ratio in the case of ANN models and nine significantly different pairs based on hit ratio for SVM models.

The term “economic growth” refers to the increase of real gross national product or gross domestic product or per capita income. National income or else national product is usually expressed as a measure of total added value of a domestic economy known as gross domestic product (GDP). Generally, GDP measures the value of economic activity within a country during a specific time period. The current study aims to find the most suitable model that adjusts on a time-series data set using Box-Jenkins methodology and to examine the forecasting ability of this model. The analysis used quarterly data for Greece from the first quarter of 1995 until the third quarter of 2019. Nonlinear maximum likelihood estimation (maximum likelihood-ML) was applied to estimate the model using the Broyden-Fletcher-Goldfarb-Shanno (BFGS) algorithm while covariance matrix was estimated using the negative of the matrix of log-likelihood second derivatives (Hessian-observed). Forecasting of the time series was achieved both with dynamic as well as static procedures using all forecasting criteria.

Significant evidence in the literature points to money demand instability and therefore inaccurate forecasting. In view of this issue, this chapter seeks to use a method, innovative for money demand literature, that is, the machine learning model to predict money demand. Specifically, this chapter uses Random Forest Regression to predict money demand using monthly data in the Indian context over the period April-1996 to December-2018 using the variables usually used in literature. The chapter finds that in money demand prediction, the Random Forest Regression performs fairly well. The results are also compared to traditional models and it is found that the Random Forest Regression model has the potential to enhance the prediction of money demand over what traditional models predicts.

The purpose of this chapter is to describe and analyze the economic advantage of the geographical location of the Republic of Belarus. The current state of the Belarusian logistics system is analyzed in detail in the chapter. Thus effects of each direction of transportations are analyzed and also approaches to assessment of their cost efficiency are formulated. The factors influencing the export of transport services as well as the development of trends in the transport sector of Belarus are defined. The main directions and ways of improvement of logistics in the Republic of Belarus are described.

The promotion of low tariffs and free trade has been the underlying driver of global economic growth. The recent political developments in the United States and Great Britain calls into question, whether free trade will be supported by the governments of the industrialized world in the future. Shortly after being inaugurated in 2017, the President of the United States has repeatedly announced his plans to impose punitive tariffs on the import of foreign products in order to protect the country’s domestic economy. Besides a controversial border adjustment tax, he has frequently brought up the possibility of imposing a 35% tariff on automobile imports. The chapter aims to analyze the effects of such a tariff on trade in the automotive sector between the United States and Germany as well as on German automobile manufacturers. It takes a quantitative approach to draw a conclusion about the relationship between import tariffs on automobiles and passenger vehicle imports from Germany to the United States utilizing a fixed effects regression model based on panel data. The model finds a significant negative correlation between the examined variables, but even in a worst case scenario, German manufacturers are resilient to the predicted revenue losses caused by a tariff increase.

We investigate whether smooth robust methods for forecasting can help mitigate pronounced and persistent failure across multiple forecast horizons. We demonstrate that naive predictors are interpretable as local estimators of the long-run relationship with the advantage of adapting quickly after a break, but at a cost of additional forecast error variance. Smoothing over naive estimates helps retain these advantages while reducing the costs, especially for longer forecast horizons. We derive the performance of these predictors after a location shift, and confirm the results using simulations. We apply smooth methods to forecasts of UK productivity and US 10-year Treasury yields and show that they can dramatically reduce persistent forecast failure exhibited by forecasts from macroeconomic models and professional forecasters.

The most popular method for calculating asset prices is the Capital Asset Pricing Model (CAPM). What is the appropriate amount of years to use in the estimation and which variation of the capital asset pricing beta provides the best results? This research looks at the out-of-sample forecasting capabilities of three popular CAPM ex-post constant beta models from 2005 to 2014. A total of 11 portfolios, five from developed and six from developing markets, are used to test the amount of input years that will reduce the mispricing in both types of markets. It is found that the best beta model to use varies between developed and developing markets. Additionally, in developing markets, a shortened span of historical years improves the pricing, contrary to popular studies that use 5 to 10 years of historical data. There are many different CAPM studies implementing various betas, using different data input lengths and run in various countries. This study empirically tests the best practices for those interested in successfully using the CAPM for their basic needs, finding that overall the simple ex-post constant beta is mispriced by 0.2 (developing) to 0.3 percent (developed). It is better to use short three-year estimation windows with the market beta in developing economies and longer nine-year estimation windows with the adjusted beta in developed economies.