AI Managed US Stock Portfolio

AI Managed US Stocks Portfolio follows a systematic, data-driven investment strategy in which a machine-learning model is used to select US equities. The investment universe consists of approximately 480 large US companies drawn primarily from the S&P 500 and Nasdaq 100. For each stock, the model analyses a range of market indicators, including price trends, momentum, volatility and the stock's position relative to its 52-week high and low. The analysis also incorporates broader market and economic conditions, such as interest-rate policy, unemployment, market volatility (VIX) and consumer confidence. A self-learning model analyses approximately 25 years of historical data to identify relationships between these signals and subsequent stock-price performance. Each stock receives a score representing the model's expected return over the following quarter. Every three months, the 10 highest-scoring stocks are selected and approximately equally weighted. Positions are normally held until the next quarterly review, when the complete universe is analysed and ranked again. The model has been developed using a strict point-in-time methodology: historical simulations only use information that would have been available at each respective date, avoiding the use of future information. The strategy does not attempt to predict short-term market movements or select stocks that necessarily rise more frequently. Instead, it seeks to identify stocks with an attractive expected return profile, including the potential for stronger gains when favourable price movements occur. The objective is to combine artificial intelligence, quantitative analysis and disciplined portfolio construction in a transparent and repeatable investment process.