canli-mcp / Markets
Event study
How a stock's price reacted to events: earnings releases (8-K item 2.02), insider purchases or sales (Form 4), any 8-K, or given dates. Abnormal returns against a market model, cumulative abnormal return per event and on average, with a t test across events. For filings use event_type (with last for the most recent few), not filing dates: it starts day 0 at the first session after the SEC accepted the filing.
The tool is event_study in canli-mcp 0.22.0, free and MIT-licensed, with no account and no API key. It comes from canli-markets-mcp, which canli-mcp includes, and it is also its own server.
Arguments
| Argument | Required | Type |
|---|---|---|
symbol | yes | string |
events | no | string[] |
event_type | no | earnings|insider_buys|insider_sales|8k |
since | no | string |
last | no | integer |
window | no | any[] |
estimation_days | no | integer |
benchmark | no | string |
model | no | market|market_adjusted|mean_adjusted |
How a model reaches it
Ask in plain words. The model finds event_study with canli-mcp's find_tool, reads its arguments with describe_tool if it needs to, and runs it with run_tool. The call has this shape, with your values in place of the angle brackets:
{
"name": "event_study",
"arguments": {
"symbol": "<string>"
}
}Words that find it
Besides its name, title and description, find_tool matches a question to event_study on these words: event, study, earnings, reaction, announcement, abnormal, return, car, stock, move, after, insider, buy, sell, market, model, significance.
More
Nothing here is investment advice.