Akarsh Verma
@akarshverma
Followers
314
Following
67
Media
327
Statuses
5K
Engineer & Architect | Procrastinating YouTuber | Loves Technology, Books, Comics & Coffee | Evolving Father & Hopeful Husband | ✨✨✨ https://t.co/z96u3JKmR1
YouTube 👉
Joined September 2009
0
0
0
If you’re building or scaling ML systems, this is worth checking out.
1
0
0
🚀 Excited to share that Packt is hosting a Virtual Machine Learning Summit this July — covering everything from Applied ML Engineering to GenAI and LLMs.
1
0
0
Turnout at AWS event today #aws #awssummit2025 . wondering who is working today in Bangalore . possibly over 10k footfall today
0
0
1
12 of 12 Enjoyed this . !!. Follow @akarshverma and subscribe to to show ❤️ ❤️. Cheers . !!.
1 of 12 🚨 Ever heard of Pair Trading? I found it while exploring risk-managed strategies. It’s market-neutral — and it blew my mind 🧵. Here's a thread describing that. Check out the video and the article for a deep dive. - -
0
0
0
11 of 12 Pair trading blends math, logic, & market psychology. Not risk-free, but deeply rewarding.
1
0
0
10 of 12 📺 Watch the full guide with code + visuals here:.👉
1
0
0
9 of 12 🔥 What’s next for me:.• Kalman filters for dynamic beta.• ML to predict mean reversion.• Volatility-based sizing.
1
0
0
8 of 12 🧠 Key lessons:.• Market neutrality helps.• Co-integration > correlation.• Python = powerful customization.
1
0
0
7 of 12 Python co-integration test:.ts.coint(stockA, stockB).p < 0.05? Likely co-integrated. To model spread:.spread = stockA - (alpha + beta * stockB).Then trade when spread hits upper/lower bounds.
1
0
0
6 of 12 📈 Visualized spread, signals, PnL in Excel. Was it basic? Yes. Did it help me learn deeply? 100%. From Excel ➡️ Python 🐍.Why?.• Handle more data.• Faster loops.• Statistical testing.• Custom logic.
1
0
0
5 of 12 🧪 First sim was in Excel :. • Price spread.• Mean & std dev.• Entry/exit thresholds.• PnL logic.• Cumulative returns.
1
0
0
4 of 12 🛠 How I chose my first pair :. ✅ Same sector.✅ Charts aligned.✅ Correlation in Excel.✅ Co-integration in Python.
1
0
0
3 of 12 The strategy works if the relationship holds. That’s why correlation ≠ co-integration. Big difference. 📉 Correlation: short-term co-movement.📈 Co-integration: long-term stable relationship. Two stocks can be correlated but not co-integrated — and drift apart forever.
1
0
0
2 of 12 Pair trading = 2 related stocks. Long one, short the other. You’re trading their relationship, not market direction. 🎯 Example: KO & PEP. If KO moves too far up vs PEP, short KO, long PEP. Bet is: the spread reverts to the mean. Profit on convergence.
1
0
0
1 of 12 🚨 Ever heard of Pair Trading? I found it while exploring risk-managed strategies. It’s market-neutral — and it blew my mind 🧵. Here's a thread describing that. Check out the video and the article for a deep dive. - -
1
0
0
The Chinese symbol for crisis 危机 is composed of two elements: one signifies danger and the other opportunity.
0
0
1