Purpose: Stock markets are essential for households for wealth creation and for firms for raising financial resources for capacity expansion and growth. Market participants, therefore, need an understanding of stock price movements. Stock market indices and individual stock prices reflect the macroeconomic environment and are subject to external and internal shocks. It is important to disentangle the impact of macroeconomic shocks, market uncertainty and speculative elements and examine them separately for prediction. To aid households, firms and policymakers, the paper proposes a granular decomposition-based prediction framework for different time periods in India, characterized by different market states with varying degrees of uncertainty.Design/methodology/approach: Ensemble empirical mode decomposition (EEMD) and fuzzy-C-means (FCM) clustering algorithms are used to decompose stock prices into short, medium and long-run components. Multiverse optimization (MVO) is used to combine extreme gradient boosting regression (XGBR), Facebook Prophet and support vector regression (SVR) for forecasting. Application of explainable artificial intelligence (XAI) helps identify feature contributions.Findings: We find that historic volatility, expected market uncertainty, oscillators and macroeconomic variables explain different components of stock prices and their impact varies with the industry and the market state. The proposed framework yields efficient predictions even during the COVID-19 pandemic and the Russia–Ukraine war period. Efficiency measures indicate the robustness of the approach. Findings suggest that large-cap stocks are relatively more predictable.Research limitations/implications: The paper is on Indian stock markets. Future work will extend it to other stock markets and other financial products.Practical implications: The proposed methodology will be of practical use for traders, fund managers and financial advisors. Policymakers may find it useful for assessing the impact of macroeconomic shocks and reducing market volatility.Originality/value: Development of a granular decomposition-based forecasting framework and separating the effects of explanatory variables in different time scales and macroeconomic periods.

Technological advancement in smart cities can have adverse effects on the environment. Timely monitoring of smart cities to preserve environmental sustainability is a thrust area of research. It can be done by using change detection with multi-temporal satellite data. The success of these methods solely depends on the calibre of the backend image segmentation and Land-use Land-cover classification technique. The limitation of using cutting-edge classification algorithms is the availability of a proper dataset and identification of the edge structure of different LULC classes. In contrast, a segmentation algorithm cannot detect LULC classes automatically. In this research, we eliminated these shortcomings by considering a hybrid approach. We proposed a multi-class Support Vector Machine (SVM) and ISODATA-embedded large-scale change detection method. This method can automatically segment, detect, and perform LULC change analysis. We have considered the multi-sensor dataset of Barasat, West Bengal, India, obtained from the WorldView satellite sensor for the experimental study. The proposed method is validated concerning three different cutting-edge methods.