This paper analyzes the market microstructure of the European Climate Exchange, the largest EU ETS trading venue. The ECX captures 2/3 of the screen traded market in EUA and more than 90% in CER. Volume growth has averaged 277% in EUA between 2005 and 2009 and 724% in CER since 2007. Spreads range from €0.0188 to €0.0406 for EUA and €0.0276 to €0.0541 for CER. The median proportion of the spread due to adverse selection reaches 76% for EUA and 75% for CER. Realized volatility, bid-ask spreads and adverse selection costs decline with verified emission releases. Market impact estimates imply that an average trade will move the EUA market by 1.05 euro centimes and the CER market 1.44. The ECX is providing between 75% and 88% of price discovery for EUA trading and between 64% and 72% for CER. We find imbalances in the order book help predict returns for up to three days. A simple trading strategy that enters the market long or short when the order imbalance is strong is profitable even after accounting for spreads and market impact.

The upward bias of the widely used Thompson–Waller estimator has been pointed out in the literature. In contrast, the current article provides a case the estimator would have downward bias: frequent continuous arrivals of orders in the same side associated with a small price change. The upward bias might be cancelled out by downward bias, and the estimator might perform better than the other methods such as Wang–Yau–Baptiste used by the U.S. Commodity Futures Trading Commission. The high-frequency data of the emissions market allows us to provide an empirical evidence.

This study applies intermarket analysis to forecast the New Zealand NZX 50 stock returns using a hybrid framework that combines econometric models with machine learning (ML) algorithms. Daily return data from 3 January 2001 to 31 December 2024 are employed to examine the interconnectedness between the New Zealand equity market and major global financial markets. This study follows a three-stage methodology: ML-based feature selection using LASSO, ridge, and elastic net regressions; econometric validation through a Global Vector Autoregressive (GVAR) model; and forecasting implementation using Support Vector Regression (SVR), Random Forest (RF), Long Short-Term Memory, and Artificial Neural Networks. Feature selection consistently identifies the Australian ASX 200, Japanese Nikkei 225, U.S. S&P 500, and U.S. 10-year Treasury yield as the most influential predictors, with Australia exerting the strongest impact. GVAR results reveal significant short-term spillover effects, but no long-term co-integrating relationships, indicating independent market trends. The U.S. market emerges as the dominant transmitter of shocks, while New Zealand acts as a net receiver. Forecasting results show RF and SVR outperform alternative models, with optimal performance achieved using a fifth-lag structure. The findings support short-term, spillover-based investment strategies and contribute to a transparent, replicable framework for forecasting equity markets in small open economies.