FinFeedAPI Blog - How to Combine SEC Filings, Stock OHLCV, and Prediction Market Data for Event-Driven Research

A company files an 8-K.

The stock moves 4%.

A prediction market tied to the same company or economic event moves from 42% to 61%.

Looking at any one of those signals tells only part of the story. Event-driven research becomes more interesting when you connect what happened, how the stock market reacted, and how expectations changed.

FinFeedAPI makes it possible to combine these three layers using its SEC API, historical Stock API, and Prediction Markets API within one developer-focused data ecosystem.

Three Data Layers, Three Different Questions

For event-driven analysis, each dataset has a different job.

Data layer	What it tells you
SEC filings	What the company disclosed
Stock OHLCV	How the equity market reacted
Prediction markets data	How expectations about an outcome changed

The value comes from putting them on the same timeline.

Instead of simply asking “Did the stock go up after the filing?”, you can ask:

Did expectations change before the filing? Did the disclosure confirm something prediction markets had already started pricing? Did the stock react more strongly than the probability market? Did trading volume or liquidity change around the same event?

That creates a much richer event dataset.

Step 1: Start With the SEC Filing

SEC filings provide the event anchor.

FinFeedAPI's SEC API provides structured access to EDGAR data, including filing metadata, original documents, full-text search, section extraction, and XBRL conversion to structured JSON.

That means researchers can build workflows around events such as:

new 8-K filings;
earnings or guidance updates in a 10-Q;
changes to risk factors in a 10-K;
mentions of mergers, cybersecurity incidents, material weaknesses, layoffs, or going-concern language.

Full-text search can identify relevant filings, while extraction endpoints can isolate specific sections rather than forcing a model or research pipeline to process the entire document.

The important output is an event record containing fields such as company, ticker, CIK, form type, filing timestamp, accession number, event type, and relevant disclosure text.

That timestamp becomes the center of the analysis.

Step 2: Measure the Stock Market Reaction

Once the event is identified, pull stock data around the same window.

FinFeedAPI provides historical stock OHLCV through:

GET /v1/ohlcv/exchange-symbol/{exchange_id}/{symbol_id}/history

The endpoint supports a requested period together with time_start, time_end, and limit, returning open, high, low, close, volume, and trade-count data in ascending time order.

A researcher could examine windows such as:

[-5 trading days, +5 trading days]

or, for shorter-term analysis:

[-1 hour, +6 hours]

From the resulting OHLCV data, you can calculate signals such as post-filing return, volume spikes, and intraperiod volatility.

Now you know how the equity market reacted.

But that still doesn't tell you how expectations about the underlying event evolved.

Step 3: Add Prediction Markets Data

Prediction markets introduce another dimension: expectations.

FinFeedAPI's Prediction Markets API provides market discovery, latest activity, historical and latest OHLCV, and current order books.

Suppose you're researching an acquisition.

The SEC filing tells you what the company disclosed.

Stock OHLCV tells you how investors repriced the company's shares.

A relevant prediction market might tell you how the perceived probability of the acquisition being completed changed around the same announcement.

Historical prediction-market OHLCV is available through:

GET /v1/ohlcv/{exchange_id}/{market_id}/history

For contracts where price represents implied probability, the close price can be used to track how market expectations changed over time.

The result is a third time series that can be aligned with the SEC event and stock data.

Put Everything on One Timeline

This is where the datasets become much more useful.

Imagine a simplified event:

09:00 — prediction-market probability: 38%
10:12 — company publishes an 8-K
10:30 — stock rises 3.1%
11:00 — prediction-market probability: 57%

The important observation isn't simply that two markets moved.

You can measure when they moved, how much they moved, and whether one market started repricing the event before the other.

A basic research pipeline can therefore look like:

SEC event detection → event timestamp → stock OHLCV window → prediction-market window → aligned event dataset

This workflow is directly supported by the filing, OHLCV, and prediction-market endpoints available through FinFeedAPI.

What Can You Measure?

Once the three sources are aligned, the interesting part starts.

You can calculate stock returns before and after a disclosure and compare them with changes in prediction-market probability.

You can look for unusual stock volume at the same time prediction-market activity increases.

You can compare the bid/ask spread before and after a filing to see whether uncertainty or liquidity changed.

FinFeedAPI also exposes current prediction-market activity with the latest trade and quote, including bid, ask, and volume information. Current order books can add another layer by showing available bids, asks, liquidity, and depth.
That creates research signals such as:

Disclosure → probability change → stock reaction

or:

Probability change → SEC disclosure → stock reaction

That second pattern can be particularly interesting when studying whether markets appeared to anticipate an event.

Example: Earnings and Guidance

Consider a company approaching an earnings release.

A prediction market could be tracking an outcome connected to earnings, economic conditions, regulatory approval, or another event affecting the business.

Before the filing, you track the probability over time.

Then a 10-Q or 8-K appears.

The SEC API provides the filing and relevant disclosure text. The Stock API measures the subsequent share-price and volume response. The Prediction Markets API shows whether expectations moved before, during, or after the disclosure.

Instead of one dataset, you now have a structured event record combining:

what management said + what the stock did + what the market expected.

From Research to Data-Driven Strategies

This doesn't mean that combining more datasets automatically produces a profitable trading signal.

But it does create better inputs for testing hypotheses.

Researchers can investigate whether certain filing language historically produces larger equity reactions, whether prediction markets react faster to specific event types, whether large probability changes precede stock moves, or whether disagreement between markets contains useful information.

This is where making money with data becomes a research problem rather than a promise.

The goal is not to assume that one signal predicts another. It is to collect enough historical events to test whether a relationship exists, how stable it is, and whether it survives transaction costs and changing market conditions.

One Research Dataset Instead of Three Separate Pipelines

Normally, this type of project means connecting separately to SEC EDGAR, a stock market data provider, and several prediction-market venues.

FinFeedAPI reduces that integration work.

The SEC API provides the disclosure layer. The Stock API provides historical OHLCV for measuring market reaction. The Prediction Markets API provides market prices, OHLCV, activity, and order books for measuring changing expectations.

The real opportunity isn't any individual API.

It's what becomes possible when you connect the datasets: what happened, what investors did, and what the market believed… on the same timeline.

Explore FinFeedAPI APIs
Start with $25 in Free Credits

How to Combine SEC Filings, Stock OHLCV, and Prediction Market Data for Event-Driven Research