Search results

This paper aims to analyze the role and performance of different artificial intelligence (AI) algorithms in forecasting future movements in the main indices of the world’s largest stock exchanges.

Drawing on finance-based theory, an empirical and experimental study was carried out using four AI-based models. The investigation comprised training, testing and analysis of model performance using accuracy metrics and F1-Score on data from 34 indices, using 9 technical indicators, descriptive statistics, Shapiro–Wilk, Student’s t and Mann–Whitney and Spearman correlation coefficient tests.

All AI-based models performed better than the markets' return expectations, thereby supporting financial, strategic and organizational decisions. The number of days used to calculate the technical indicators enabled the development of models with better performance. Those based on the random forest algorithm present better results than other AI algorithms, regardless of the performance metric adopted.

The study expands knowledge on the topic and provides robust evidence on the role of AI in financial analysis and decision-making, as well as in predicting the movements of the largest stock exchanges in the world. This brings theoretical, strategic and managerial contributions, enabling the discussion of efficient market hypothesis (EMH) in a complex economic reality – in which the use of automation and application of AI has been expanded, opening new avenues of future investigation and the extensive use of technical analysis as support for decisions and machine learning.

The AI algorithms' flexibility to determine their parameters and the window for measuring and estimating technical indicators provide contextually adjusted models that can entail the best possible performance. This expands the informational and decision-making capacity of investors, managers, controllers, market analysts and other economic agents while emphasizing the role of AI algorithms in improving resource allocation in the financial and capital markets.

The originality and value of the research come from the methodology and systematic testing of the EMH through the main indices of the world’s largest stock exchanges – something still unprecedented despite being widely expected by scholars and the market.

The purpose of this study is to examine the buyers’ preferences influencing the purchase of privately developed affordable housing in Kolkata and to determine whether unsold houses result from misalignment with these preferences.

The literature review and user-opinion survey identified 119 independent variables that indicate buyers’ preferences. A questionnaire survey of 383 households in affordable housing units from 32 housing complexes in Kolkata recorded buyers’ preferences and satisfaction against the independent variables grouped under five levels of characteristics. The product weights of variables derived from the rank sum method and percentage satisfaction give the Utility Score. Multivariate regression and univariate linear regressions were conducted to determine the significance of each Level of characteristics and each variable, identifying the significant variables that would affect the sale of affordable houses.

The multivariate regression analysis has indicated that 68.56% of the variation in the percentage of unsold houses was explained by the five utility scores, which affirms that misalignment with buyers’ preferences significantly affects the sale of privately developed affordable houses. Furthermore, building and neighbourhood-level utility show the highest significance as predictors, while city-level and miscellaneous utility have moderate significance, but housing complex-level utility lacks statistical significance.

This study addresses a research gap in privately developed affordable housing in Kolkata, enhancing understanding of buyer preferences in this segment.

The current approach to valuing customers is based on the notion of discounted profit generated by the customers over the lifetime of the relationship, also known as customer lifetime value (CLV). However, in the financial services industry, the customers who contribute the most to the profitability of a firm are also the riskiest customers. If the riskiness of a customer is not considered, firms will overestimate the true value of that customer. This paper proposes a methodology to adjust CLV for different types of risk factors and creates a comprehensive measure of risk-adjusted lifetime value (RALTV).

Using data from a major credit card company, we develop a measure of risk adjusted lifetime value (RALTV) that accounts for diverse types of customer risks. The model is estimated using Stochastic Frontier Analysis (SFA).

Major findings indicate that rewards cardholders and affinity cardholders tend to score higher within the RALTV framework than non-rewards cardholders and non-affinity cardholders, respectively. Among the four different modes of acquisition, the Internet generates the highest RALTV, followed by direct mail.

This paper not only controls for different types of consumer risks in the financial industry and creates a comprehensive risk-adjusted lifetime value (RALTV) model but also shows empirically the value of using RALTV over CLV for predicting future performance of a set of customers. Further, we investigate the impact of a firm’s acquisition and retention strategies on RALTV. The measure of risk-adjusted lifetime value is invaluable for managers in financial services.