Skip to main content

​
Analyze Insider Trading Patterns

Corporate insiders (officers, directors, and 10% shareholders) must report their trades within two business days on SEC Form 4. This tutorial shows how to retrieve insider transaction data, filter by transaction type, and identify patterns.

​
Prerequisites

  • An Omni Datastream API key (set as OMNI_DATASTREAM_API_KEY)
  • Basic familiarity with REST APIs
  • (Optional) Python 3.8+ or Node.js 18+ for SDK examples

​
Step 1 — Get insider transactions for a company

Use /v1/insiders to retrieve recent insider transactions for a given ticker.

​
curl

curl -H "x-api-key: $OMNI_DATASTREAM_API_KEY" \
  "https://api.secapi.ai/v1/insiders?ticker=AAPL&limit=10"

​
Python

from omni_datastream_py import OmniDatastreamClient

client = OmniDatastreamClient(api_key="your-api-key")

transactions = client.insiders.list(ticker="AAPL", limit=10)

for txn in transactions.data:
    direction = "BUY" if txn.transaction_type == "purchase" else "SELL"
    print(f"{txn.transaction_date} | {txn.insider_name} ({txn.insider_title})")
    print(f"  {direction}: {txn.shares:,} shares @ ${txn.price:.2f} = ${txn.value:,.0f}")
    print()

​
JavaScript

import { OmniDatastreamClient } from "@omni-datastream/sdk-js";

const client = new OmniDatastreamClient({
  apiKey: process.env.OMNI_DATASTREAM_API_KEY!,
});

const transactions = await client.insiders.list({
  ticker: "AAPL",
  limit: 10,
});

for (const txn of transactions.data) {
  const direction = txn.transactionType === "purchase" ? "BUY" : "SELL";
  console.log(`${txn.transactionDate} | ${txn.insiderName} (${txn.insiderTitle})`);
  console.log(
    `  ${direction}: ${txn.shares.toLocaleString()} shares @ $${txn.price.toFixed(2)} = $${txn.value.toLocaleString()}`
  );
  console.log();
}

​
Expected output

2024-12-10 | Tim Cook (Chief Executive Officer)
  SELL: 200,000 shares @ $248.50 = $49,700,000

2024-12-05 | Jeff Williams (Chief Operating Officer)
  SELL: 100,000 shares @ $243.20 = $24,320,000

2024-11-20 | Deirdre O'Brien (SVP, Retail + People)
  SELL: 50,000 shares @ $231.45 = $11,572,500

​
Step 2 — Filter for insider purchases

Insider purchases are often considered more meaningful than sales (insiders sell for many reasons, but they buy for only one). Filter by transaction type.

​
curl

curl -H "x-api-key: $OMNI_DATASTREAM_API_KEY" \
  "https://api.secapi.ai/v1/insiders?ticker=JPM&transaction_type=purchase&limit=10"

​
Python

buys = client.insiders.list(
    ticker="JPM",
    transaction_type="purchase",
    limit=10,
)

total_value = sum(txn.value for txn in buys.data)
print(f"Recent insider purchases at JPM: {len(buys.data)} transactions")
print(f"Total value: ${total_value:,.0f}")
print()

for txn in buys.data:
    print(f"  {txn.transaction_date} | {txn.insider_name}: {txn.shares:,} shares (${txn.value:,.0f})")

​
JavaScript

const buys = await client.insiders.list({
  ticker: "JPM",
  transactionType: "purchase",
  limit: 10,
});

const totalValue = buys.data.reduce((sum, txn) => sum + txn.value, 0);
console.log(`Recent insider purchases at JPM: ${buys.data.length} transactions`);
console.log(`Total value: $${totalValue.toLocaleString()}`);
console.log();

for (const txn of buys.data) {
  console.log(
    `  ${txn.transactionDate} | ${txn.insiderName}: ${txn.shares.toLocaleString()} shares ($${txn.value.toLocaleString()})`
  );
}

​
Step 3 — Scan for cluster buying across a sector

When multiple insiders at different companies in the same sector start buying, it can signal sector-wide conviction. Scan a list of tickers for recent insider purchases.

​
Python

BANK_TICKERS = ["JPM", "BAC", "WFC", "C", "GS", "MS"]

print("Sector Insider Buy Scan - Major Banks")
print("=" * 60)

for ticker in BANK_TICKERS:
    buys = client.insiders.list(
        ticker=ticker,
        transaction_type="purchase",
        date_from="2024-10-01",
        limit=5,
    )

    if buys.data:
        total = sum(t.value for t in buys.data)
        print(f"\n{ticker}: {len(buys.data)} insider buys (${total:,.0f} total)")
        for txn in buys.data:
            print(f"  {txn.transaction_date} {txn.insider_name}: ${txn.value:,.0f}")
    else:
        print(f"\n{ticker}: No insider purchases since 2024-10-01")

​
Expected output

Sector Insider Buy Scan - Major Banks
============================================================

JPM: 3 insider buys ($4,250,000 total)
  2024-12-01 Jamie Dimon: $3,000,000
  2024-11-15 Mary Erdoes: $750,000
  2024-10-20 Daniel Pinto: $500,000

BAC: 1 insider buys ($250,000 total)
  2024-11-10 Brian Moynihan: $250,000

WFC: No insider purchases since 2024-10-01

C: 2 insider buys ($1,100,000 total)
  2024-12-05 Jane Fraser: $800,000
  2024-10-30 Mark Mason: $300,000

​
Step 4 — Track an individual insider over time

Focus on a single insider to understand their trading history.

​
curl

curl -H "x-api-key: $OMNI_DATASTREAM_API_KEY" \
  "https://api.secapi.ai/v1/insiders?ticker=JPM&insider_name=Jamie+Dimon&limit=20"

​
Python

history = client.insiders.list(
    ticker="JPM",
    insider_name="Jamie Dimon",
    limit=20,
)

buys = [t for t in history.data if t.transaction_type == "purchase"]
sells = [t for t in history.data if t.transaction_type == "sale"]

print(f"Jamie Dimon - JPM Trading History")
print(f"  Purchases: {len(buys)} ({sum(t.shares for t in buys):,} shares)")
print(f"  Sales:     {len(sells)} ({sum(t.shares for t in sells):,} shares)")

​
Next steps

  • Combine with 13F data: Cross-reference insider trades with institutional holdings changes for a complete ownership picture.
  • Set up alerts: Use the Filing Monitor tutorial to get notified when new Form 4 filings are published.
  • Build scoring models: Weight insider purchases by position (CEO buys often carry more signal than director buys) and transaction size relative to holdings.
See the API Reference for the full insider transactions endpoint specification.