Market Strategy Prompts: Synthesizing an Equity Market Outlook
AI generated Beta test version
Last updated: 17 September 2025
Objective:
Synthesize implications across the recent trends in a broad, economy-wide representative set of macroeconomic indicators, geopolitical and fiscal trends, with market related indicators (rates, liquidity, spreads, earnings, valuations, prices and volatility) to generate balanced outlook scenarios for the equity market over a 3-month horizon.
Explanation:
Market commentary often falls into one of two traps: it either buries the reader in a flood of raw data without clear implications, or it leaps straight to bold forecasts without showing the evidence. What actually helps investors navigate the next few months, and also challenges any preconceptions they might have, is a disciplined synthesis of macroeconomic signals and market indicators - a structured way of asking, what story is the hard and soft data in totality really telling us?
This style of analysis starts by laying out the building blocks: growth, inflation, labor markets, consumer and business sentiment, production and geopolitics. It then links these fundamentals to the plumbing of markets — interest rates, yield curves, credit spreads, earnings revisions, valuations and price and volatility trends. Instead of treating each dataset in isolation, it asks how they interact: Does a weakening labor market soften inflation pressure but undermine consumer spending? Does a steepening yield curve reflect policy easing or rising long-term risk premia? Are equity valuations assuming resilience even as surveys and credit begin to turn down?
The value of this integrated approach is twofold. First, it helps identify when the market’s pricing diverges from the macro trajectory - signaling either opportunity or risk. Second, it forces transparency: readers can see each data series, the forward expectations, and how they roll up into a three-month outlook with probabilities, not just a single “call.”
In what follows, we’ll walk through exactly that process. Use this prompt set to review the latest country specific macro data, track financial conditions, assess earnings trends, and then stitch these inputs together into a clear short-term equity market outlook. The goal isn’t to predict with certainty - it’s to provide a coherent framework so investors can create their own analysis to help understand what’s priced in, what’s shifting, and where the risks and opportunities really lie.
Link to blog post explanation:
Using AI to Construct Market Scenarios
Preferred Model(s):
Dual model technique recommended. Gemini 2.5+ and ChatGPT-5+ with Chain-of-Verification (CoVe) follow up in each. See execution notes
Important Execution Notes:
The prompt requires the user to upload an extensive dataset of weekly closing prices for the main country index being examined (e.g. S&P 500 index). I use 25 years of weekly closes to ensure it covers market cycles.
Customize the prompt for the country (currently set for the US).
As this style of analysis calls for an outlook to be created, there is wide latitude for the Gen AI model to make conclusions across a broad range of potential outcomes. We wish to ensure those conclusions are well reasoned, anchored in objective data and proper weighting is applied to each step through the analysis to the synthesized conclusion. For this we use a 2 model technique with Chain-of-Verification follow up steps to interrogate the analysis.
The primary prompt is run through BOTH Gemini and ChatGPT to elicit initial responses from each model.
Add CoVe steps: Each model is then asked to generate a list of questions to assess the quality and depth of the response and the internal consistency of the conclusions drawn.
The investor should also append any questions of their own to clarify the analysis or implications drawn in the various steps, to the list of verification questions. This can include additional considerations that spring from the analysis or clarifications of data or analysis that appears unusual.
Each model is then asked to answer those combined questions, and on the basis of that, update its scenarios and conclusions as to the market outlook.
The use of two models allows the investor to bring considerations that seem pertinent to the attention of the other model and vice versa or challenge reasoning the model appears to be using in a step.
Don’t be surprised to potentially see one model gradually change its views and conclusions via the verification steps. This is desirable as it is strengthening the reasoning, conclusions and confidence the investor can hold in the combined output. It is not uncommon to see one model hold its line through the verification while the other, starting from an alternative point of view gradually shifts via interrogation close to the other view. This shift is valuable information in itself.
Quantitative conclusions and even directionality can differ between models at the conclusion but this offers another valuable viewpoint. The models should essentially provide the same data and calculate the same trends - the differences will lie in the reasoned interpretations.
The investor should isolate what interpretations both models agree on and where they differ. Then look for why those reasonings differ (eg weighting of different datasets in the conclusion), weighting of hard data vs soft data or current trend versus a more forward looking (& risky) prediction. The investor can then make their own informed conclusion where they sit in this range.
Sample Output:
Copy/Paste Prompt Set:
Important note: Subscribers can use this prompt set for their own analysis. However, the prompt is copyrighted by The Inferential Investor, paywalled, and must not be shared without permission.