Systematic Investment Planning

𝗜𝗻𝗱𝗶𝗮’𝘀 𝗲𝘃𝗼𝗹𝘃𝗶𝗻𝗴 𝗙𝗗𝗜 𝗿𝗲𝗴𝗶𝗺𝗲 𝗶𝘀 𝘀𝗵𝗮𝗽𝗶𝗻𝗴 𝘁𝗵𝗲 𝗳𝘂𝘁𝘂𝗿𝗲 𝗼𝗳 𝗮𝗹𝘁𝗲𝗿𝗻𝗮𝘁𝗶𝘃𝗲 𝗶𝗻𝘃𝗲𝘀𝘁𝗺𝗲𝗻𝘁𝘀 — 𝗵𝗲𝗿𝗲’𝘀 𝘄𝗵𝗮𝘁 𝗔𝗜𝗙 𝗺𝗮𝗻𝗮𝗴𝗲𝗿𝘀 𝗻𝗲𝗲𝗱 𝘁𝗼 𝗸𝗻𝗼𝘄. Alternative Investment Funds (AIFs) continue to attract significant interest from foreign investors — and rightly so. However, with this capital inflow comes a heightened need for regulatory vigilance, especially when AIFs receiving foreign investment undertake downstream investments in Indian entities. To ensure transparency and control, the Reserve Bank of India (RBI), under the Foreign Exchange Management (Non-Debt Instruments) Rules, 2019, has laid down a robust reporting framework. Two key filings in this compliance matrix are Form DI and Form InVi: 𝗙𝗼𝗿𝗺 𝗗𝗜 – 𝗗𝗼𝘄𝗻𝘀𝘁𝗿𝗲𝗮𝗺 𝗜𝗻𝘃𝗲𝘀𝘁𝗺𝗲𝗻𝘁𝘀 When a foreign-owned and controlled AIF (FOCC AIF) makes a downstream investment (such as equity shares, CCPS, or CCDs) in an Indian entity, it constitutes indirect foreign investment. (a) Form DI must be filed on the FIRMS portal within 30 days from the date of allotment of such instruments by the Indian investing entity. (b) Compliance with India’s FDI policy — including sectoral caps, pricing guidelines, and prior approvals where applicable — becomes mandatory. 𝗙𝗼𝗿𝗺 𝗜𝗻𝗩𝗶 – 𝗜𝗻𝗱𝗶𝗿𝗲𝗰𝘁 𝗙𝗼𝗿𝗲𝗶𝗴𝗻 𝗜𝗻𝘃𝗲𝘀𝘁𝗺𝗲𝗻𝘁 𝘃𝗶𝗮 𝗜𝗻𝘃𝗲𝘀𝘁𝗺𝗲𝗻𝘁 𝗩𝗲𝗵𝗶𝗰𝗹𝗲𝘀 Where a foreign investor invests in an Indian AIF, and the AIF subsequently invests in an Indian entity, Form InVi comes into play. (a) The AIF must report this indirect foreign investment within 30 days of issuing units or allocating investments to the foreign investor. 𝗔𝗱𝗱𝗶𝘁𝗶𝗼𝗻𝗮𝗹 𝗖𝗼𝗺𝗽𝗹𝗶𝗮𝗻𝗰𝗲 𝗥𝗲𝗾𝘂𝗶𝗿𝗲𝗺𝗲𝗻𝘁𝘀 AIFs with foreign investment must also navigate: (a) KYC norms & beneficial ownership disclosures under PMLA (b) Custodian oversight when a single investor (or group) contributes ≥ 50% of corpus (c) SEBI & RBI scrutiny on sectoral limits, pricing, and entry routes Delays or lapses in Form DI or InVi filings may trigger penal consequences under FEMA, additional regulatory scrutiny, and potential disqualification from AIF regime benefits, potentially impairing future capital raises. As India sharpens its focus on foreign capital, transparency and compliance are the cornerstones of trust. For FOCC AIFs, timely and accurate reporting under Form DI and Form InVi is not just a regulatory requirement, but it’s central to safeguarding credibility in India’s alternative investment ecosystem. Neha Londhe I ANB Legal #AlternativeInvestments #ForeignInvestment #RegulatoryCompliance

📈💼𝐇𝐞𝐫𝐞 𝐚𝐫𝐞 𝐭𝐡𝐫𝐞𝐞 𝐦𝐚𝐣𝐨𝐫 𝐚𝐧𝐚𝐥𝐲𝐭𝐢𝐜𝐬 𝐭𝐞𝐜𝐡𝐧𝐢𝐪𝐮𝐞𝐬 𝐭𝐡𝐚𝐭 𝐞𝐯𝐞𝐫𝐲 𝐟𝐢𝐧𝐚𝐧𝐜𝐞 𝐩𝐫𝐨𝐟𝐞𝐬𝐬𝐢𝐨𝐧𝐚𝐥 𝐬𝐡𝐨𝐮𝐥𝐝 𝐤𝐧𝐨𝐰, 𝐚𝐥𝐨𝐧𝐠 𝐰𝐢𝐭𝐡 𝐞𝐱𝐚𝐦𝐩𝐥𝐞𝐬 𝐨𝐟 𝐰𝐡𝐞𝐫𝐞 𝐭𝐡𝐞𝐲 𝐚𝐫𝐞 𝐚𝐩𝐩𝐥𝐢𝐞𝐝: 1️⃣ 𝐌𝐮𝐥𝐭𝐢𝐩𝐥𝐞 𝐊𝐢𝐧𝐝𝐬 𝐨𝐟 𝐑𝐞𝐠𝐫𝐞𝐬𝐬𝐢𝐨𝐧: Regression analysis is a fundamental statistical technique that helps us understand the relationship between variables. In finance, it's widely used in Asset Pricing models. For instance, the Capital Asset Pricing Model (CAPM) relies on regression to determine the expected return of an asset based on its beta and the market risk premium. Mastering various regression types can help you uncover hidden patterns and factors affecting asset prices. 2️⃣ 𝐓𝐢𝐦𝐞 𝐒𝐞𝐫𝐢𝐞𝐬 𝐀𝐧𝐚𝐥𝐲𝐬𝐢𝐬: Time series analysis is essential for forecasting and understanding the dynamics of financial data over time. It plays a crucial role in price forecasts and technical analysis. Traders and analysts often use time series techniques to identify trends, seasonality, and cycles in historical price data. Being proficient in this area allows you to make informed decisions based on historical patterns. 3️⃣ 𝐀𝐑𝐈𝐌𝐀 𝐌𝐨𝐝𝐞𝐥𝐬 (𝐀𝐮𝐭𝐨𝐑𝐞𝐠𝐫𝐞𝐬𝐬𝐢𝐯𝐞 𝐈𝐧𝐭𝐞𝐠𝐫𝐚𝐭𝐞𝐝 𝐌𝐨𝐯𝐢𝐧𝐠 𝐀𝐯𝐞𝐫𝐚𝐠𝐞): ARIMA models are a powerful tool for forecasting dependent variables in finance, especially when dealing with multi-timeframe non-panel data dependent on multiple independent variables. These models can help you make predictions about future financial variables, such as stock prices or economic indicators, by capturing both short-term and long-term trends. Learning these analytics techniques can be a game-changer in finance. They enable you to: ✅ 𝐔𝐧𝐜𝐨𝐯𝐞𝐫 𝐑𝐞𝐥𝐚𝐭𝐢𝐨𝐧𝐬𝐡𝐢𝐩𝐬: By using regression analysis, you can identify key drivers behind financial outcomes, allowing for better decision-making. ✅ 𝐀𝐜𝐜𝐮𝐫𝐚𝐭𝐞 𝐅𝐨𝐫𝐞𝐜𝐚𝐬𝐭𝐢𝐧𝐠: Time series and ARIMA models enhance your ability to forecast prices, risk, and returns accurately, which is vital in the ever-changing financial landscape. ✅ 𝐄𝐯𝐚𝐥𝐮𝐚𝐭𝐞 𝐏𝐨𝐬𝐢𝐭𝐢𝐨𝐧𝐬: Armed with these techniques, you can evaluate your investment positions more effectively, helping you manage risk and optimize returns. So, if you're in finance, consider investing your time in mastering these analytics techniques. They're your artillery for making informed decisions, managing risk, and achieving success in the world of finance. 📊📉💰 #FinanceAnalytics #DataDrivenDecisions #InvestingWisdom #businessanalytics #dataanalysis #learningandgrowing

Compliance in private markets is not a one time checkbox. It is an ongoing obligation that compounds with every new investment you make. Every deal you close triggers a fresh set of requirements: • KYC verification • PAN checks • GSTIN validation • DIN lookups • Regulatory filings such as PAS-3 and MGT-14 • Audit trails that need to be maintained for years And compliance does not end once the investment is made. The same records become critical during follow-on rounds, exits, audits, due diligence reviews, and regulatory inspections. What starts as manageable with a handful of investments quickly becomes difficult to sustain as a portfolio scales, with multiple timelines and dependencies running in parallel. India's private market ecosystem is maturing. SEBI's oversight is expanding, and institutional capital entering the space is asking sharper questions about how portfolios are being managed. Informal recordkeeping that worked five years ago is becoming a liability. At nucleo, compliance is built into the investment workflow. Every verification, filing, and audit trail is automated and maintained continuously, helping investors stay compliant without the operational overhead that typically comes with it. The cost of getting compliance right is time. The cost of getting it wrong is everything that follows.

Asset pricing models use various variables to forecast future returns of assets like stocks. These models help investors identify potentially high-performing assets. Assets are interconnected through factors like supply chains, industry sectors, and market conditions, influencing their relative prices. Graph networks are well-suited for modeling these complex relationships. Existing GNN-based asset price prediction models often focus on fixed asset groups and static relationships, neglecting the dynamic nature of asset pools and their interconnections. As financial markets are dynamic, models must adapt to changes like new market entries, asset maturation, and corporate events. This requires a flexible framework that can adapt to the dynamic nature of asset pools and their interconnections. To address the dynamic nature of the market for asset pricing, the authors of [1] propose DySTAGE (Dynamic-graph-representation-learning via Spatio-Temporal Attention and Graph Encodings), a framework with a universal formulation that transforms asset pricing time series into dynamic graphs, accommodating the addition, deletion, and changes in correlations of assets which includes a graph learning model specifically designed for this purpose. In the DySTAGE framework, assets at various historical time steps are structured as a sequence of dynamic graphs, where connections between assets reflect their long-term correlations. DySTAGE effectively captures both topological and temporal patterns. The Topological Module deploys Asset Influence Attention to learn global interrelationships among assets, further enhanced by Asset-wise Importance Encoding, Pair-wise Spatial Encoding, and Edge-wise Correlation Encoding. In the Temporal Module, DySTAGE encapsulates node representations across the temporal dimension through an attention mechanism. #QuantFinance They validate DySTAGE through extensive experiments on 3 real-world stock pricing datasets. The results show that DySTAGE outperforms popular benchmarks in return prediction and provides profitable investment strategies. The link to their paper [1] is shared in the comments.

It is incredibly lonely to carry the weight of everyone else's future on your shoulders. When you are the one in charge, you cannot show fear, even when you are panicking in silence. This heavy pressure makes mistakes feel even more dangerous, especially now as Indian companies are investing overseas more than ever - setting up subsidiaries, entering joint ventures, and acquiring businesses abroad. But the way regulators look at these investments is changing. It is no longer just about whether the paperwork was filed correctly. The bigger question is: did the company really understand what it was investing in? Who were the partners? What checks were done? Was the overseas business a genuine operating entity? And was there a clear commercial reason behind the investment? Recent developments suggest that overseas investments may receive closer attention, especially around due diligence, governance, and movement of funds. A few areas where companies could find themselves exposed: 1. Due diligence on foreign partners KYC, ownership verification, beneficial ownership checks, sanctions screening, and background diligence can become critical if questions arise later. 2. Proving commercial substance Regulators may look beyond documents and ask: • Does the entity have real operations? • Does it take independent decisions? • Does the investment have a clear business rationale? 3. Monitoring after investment Compliance does not end once funds are remitted. Companies need visibility over financial performance, related-party transactions, group fund flows, and major restructuring decisions. 4. Maintaining the decision trail Years later, the key question may be: “Why was this investment made?” Board approvals, diligence reports, evaluations, and monitoring records can become critical evidence. Having a clear record of the approvals, diligence, and reasoning behind the decision can make a significant difference. As Indian businesses continue expanding globally, the focus will likely move from simply reporting overseas investments to understanding how and why those investments were made. For companies with overseas structures, this is worth reviewing before a regulator asks the questions. --- ✍ Does your business have overseas investments? Share below!

The recent market shocks have left a tremendous effect on investors’s mindmap. The volatility and the jump in the asset prices movements are extremely high. On a behavioural finance level, there is surely panic in the market leaving less headroom to ponder about the situations for normal retail investors. Thus, the implementation of mathematical models becomes a necessity not only to predict pricing value but considering volatility, jumps and high shocks. Although, the reason is different but at the end considering the dip in Japan stock market was lower than the Covid-19 pandemic. Using stochastic process mathematical models like Heston model could be used to predict both the asset price and its volatility, allowing for a mean-reverting volatility process while Hull White model for incorporating jumps in the asset prices. This way we get the volatility, jump and asset price. Also, if we consider multivariate volatility (time varying correlations with standardized returns) with correlation b/w the multiple assets, a great recommendation to opt for the extended GARCH model with dynamic conditional coorelation (DCC). Once you could predict the dynamic correlation with varying time portfolio optimization becomes more efficient with time-varying covariance matrix. No wonder, why maths with finance using tech makes such predictions better and high accuracy rates. #quantitativefinance #quant #finance #riskmanagement #japan

📈 THE ONE MODEL EVERY LONG-TERM INVESTOR SHOULD KNOW In an environment where forecasting equity returns is both art and science, the Grinold-Kroner model stands out as a practical, conceptually elegant tool for long-term investors. At its core, the model decomposes expected equity returns into intuitive components: Expected Return ≈ Dividend Yield + Earnings Growth + Repricing (ΔP/E) – Change in Shares Outstanding This framework reminds us that returns don’t emerge from thin air - they are grounded in fundamentals: income, growth, and valuation. ✅ Why it matters: It provides a structured approach to forecast returns over the long run. It forces investors to disaggregate drivers - helping clarify assumptions and test sensitivities. It highlights the importance of valuation levels, reminding us that multiple expansion can’t sustain returns indefinitely. In today’s high-valuation, uncertain-growth environment, using models like Grinold-Kroner can bring much-needed discipline and realism to capital market expectations. Whether you're managing a pension portfolio, building an asset allocation model, or advising clients, this model offers a valuable lens. 🔍 Key insight: While no model predicts the future perfectly, frameworks like this help us ask better questions and make more grounded decisions. #Investing #GrinoldKroner 

A new study finds that transformers, the technology behind ChatGPT, can be trained directly on stock return data. Researchers trained a transformer model on up to 2 billion datapoints of return data across 34 years and 94 countries, using about 50,000 GPU hours. The researchers at Manchester, UCL, and Shanghai University say this is the first comprehensive study of Time Series Foundational Models in global markets. Time series data is any set of observations recorded in order over time (think temperature, daily sales figures, etc). 𝗛𝗲𝗿𝗲’𝘀 𝘄𝗵𝗮𝘁 𝘁𝗵𝗲𝘆 𝗱𝗶𝗱: The researchers took two popular "foundation models" for time series forecasting (Chronos from Amazon, TimesFM from Google) and asked them to predict next-day stock returns. They trained these models from scratch by using only financial data. They then compared the model they built against off-the-shelf versions trained on generic time series data. They compared all of this against simpler, well-established methods that quants already use (specifically ensemble models like gradient-boosted trees, which are basically very sophisticated decision trees). 𝗛𝗲𝗿𝗲’𝘀 𝘄𝗵𝗮𝘁 𝘁𝗵𝗲𝘆 𝗳𝗼𝘂𝗻𝗱: The off-the-shelf models flopped. When you just download these foundation models and point them at stock data, they perform terribly — worse than much simpler techniques. Fine-tuning helped a bit, but not enough to close the gap. Training from scratch worked surprisingly well. When the researchers trained these same architectures using only financial data, performance jumped dramatically. The models started generating meaningful trading signals that translated into actual portfolio returns. A Chronos model pre-trained on financial time series achieved a 36.84% annualized return and a Sharpe ratio of 5.42, compared to losses when used out of the box. The traditional quant benchmark still edged it out at 47.25% in this test, but the results suggest transformers can become more competitive with more data. The researchers released their models publicly through FinText. ai and Hugging Face, which should help with follow-on work. I asked lead author, Eghbal Rahimikia, for his big picture takeaway: 𝘚𝘤𝘢𝘭𝘪𝘯𝘨 𝘮𝘰𝘥𝘦𝘭 𝘴𝘪𝘻𝘦 𝘢𝘯𝘥 𝘦𝘹𝘱𝘢𝘯𝘥𝘪𝘯𝘨 𝘥𝘢𝘵𝘢 𝘤𝘰𝘷𝘦𝘳𝘢𝘨𝘦 𝘰𝘧𝘧𝘦𝘳 𝘢 𝘱𝘳𝘰𝘮𝘪𝘴𝘪𝘯𝘨 𝘱𝘢𝘵𝘩 𝘵𝘰𝘸𝘢𝘳𝘥 𝘪𝘮𝘱𝘳𝘰𝘷𝘪𝘯𝘨 𝘱𝘳𝘦𝘥𝘪𝘤𝘵𝘪𝘷𝘦 𝘱𝘦𝘳𝘧𝘰𝘳𝘮𝘢𝘯𝘤𝘦 𝘪𝘯 𝘢𝘴𝘴𝘦𝘵-𝘳𝘦𝘵𝘶𝘳𝘯 𝘧𝘰𝘳𝘦𝘤𝘢𝘴𝘵𝘪𝘯𝘨. 𝘏𝘰𝘸𝘦𝘷𝘦𝘳, 𝘱𝘳𝘰𝘨𝘳𝘦𝘴𝘴 𝘳𝘦𝘮𝘢𝘪𝘯𝘴 𝘤𝘰𝘯𝘴𝘵𝘳𝘢𝘪𝘯𝘦𝘥 𝘣𝘺 𝘤𝘰𝘮𝘱𝘶𝘵𝘢𝘵𝘪𝘰𝘯𝘢𝘭 𝘭𝘪𝘮𝘪𝘵𝘢𝘵𝘪𝘰𝘯𝘴 𝘢𝘯𝘥 𝘵𝘩𝘦 𝘴𝘤𝘢𝘳𝘤𝘪𝘵𝘺 𝘰𝘧 𝘭𝘢𝘳𝘨𝘦-𝘴𝘤𝘢𝘭𝘦, 𝘩𝘪𝘨𝘩-𝘲𝘶𝘢𝘭𝘪𝘵𝘺 𝘧𝘪𝘯𝘢𝘯𝘤𝘪𝘢𝘭 𝘥𝘢𝘵𝘢. Transformers, the same technology behind ChatGPT, do seem well-suited to financial forecasting. You just have to train them on lots of financial data from the start. Links to the paper and my writeup below. Eghbal Rahimikia Hao Ni Weiguan Wang

Traditionally, time series forecasting has been treated as a pure sequence modeling problem. But real-world numbers don't exist in a vacuum—they are driven by unstructured, volatile context like breaking news and global events. While specialized Time Series Foundation Models (TSFMs) excel at identifying numerical patterns, they are often blind to these critical textual signals. Our new work, Nexus, tackles this by framing forecasting as an agentic reasoning problem. We introduce a multi-agent framework that seamlessly integrates unstructured contextual information with numerical data to synthesize accurate, well-reasoned forecasts. Key Results: 📈 ✅ Strong Performance Gains: Evaluated on highly volatile datasets—including Zillow real estate metrics and stock market equities succeeding LLM knowledge cutoffs—Nexus consistently matches or outperforms state-of-the-art TSFMs and strong LLM baselines. 🧠 Multi-Agent Decomposition: Our architecture isolates macro- and micro-level temporal fluctuations into specialized reasoning stages, allowing agents to process complex, multimodal reality much like a human financial analyst would. 🚀 Explicit Reasoning Traces: Beyond just outputting a number, Nexus produces high-quality reasoning logs that explicitly show the why behind each forecast, greatly improving interpretability and trust in the system. 💡 We believe that delivering on the promise of AI agents means pushing the boundaries of how systems reason. Nexus proves that real-world forecasting extends well beyond simple numerical extrapolation. 🔗 Read the full paper here: https://lnkd.in/gSKK3_zE Authors: Sarkar Snigdha Sarathi Das, Palash Goyal, Mihir Parmar, Nanyun (Violet) Peng, Vishy Tirumalashetty, Chun-Liang Li, Rui Zhang, Jinsung Yoon, Tomas Pfister #AI #ArtificialIntelligence #MachineLearning #TimeSeries #AIagents #LLM #CloudAI #Research

Econometric Model - S&P 500 Index, Daily, Not Seasonally Adjusted (Financial Econometrics) Period: 11/08/2020 - 11/08/2025 Understanding Market Risk: A GARCH Model for S&P 500 Volatility: I’m sharing a robust volatility model using S&P 500 Index data, using an ML-estimated GARCH(1,1). This model is sound, practical, and directly linked to real-world risk management. Interpretation of the model: ▪️ Mean Equation (Expected Return) Constant = 0.000841 (p = 0.0014) → average daily gain of ~0.08% Lagged return (yₜ₋₁) is insignificant (p = 0.31), confirming that returns are nearly unpredictable day-to-day, consistent with market efficiency i.e. The lagged return’s coefficient is small and not statistically significant (p = 0.31), which means past daily returns do not help predict today’s return—exactly what the efficient-markets hypothesis implies for liquid equity indexes like the S&P 500. ▪️ Variance Equation (Volatility Dynamics) ARCH term (RESIDₜ₋₁² = 0.1295) (p < 0.0001): Yesterday’s big surprises sharply increase today’s volatility—the clustering of risk. GARCH term (σₜ₋₁ = 0.8348) (p < 0.0001): Yesterday’s forecasted risk strongly carries into today—volatility persistence. Constant = 4.47×10⁻⁶ (p < 0.0001): Baseline variance when shocks are absent. The sum of ARCH + GARCH = 0.96 (<1) assures stationary risk dynamics—volatility spikes dissipate over time, not explode. Why It Matters Risk Management: Accurately forecasting volatility is crucial for Value-at-Risk (VaR), stress testing, and option pricing. Economic Reality: In times of market turmoil (e.g., sudden macro shocks or geopolitical events), this model quantifies how “risk fires burn”—large moves beget further turbulence. Institutional Appeal: Financial institutions rely on GARCH-type models daily to allocate capital, set trading limits, and design hedging strategies. Plain Takeaway “Markets don’t move smoothly—big swings today raise the chances of big swings tomorrow, but calm periods prevail in the longer run. My GARCH model captures exactly how risk clusters and slowly fades. It's a powerful tool for managing market volatility. Source of Data: Federal Reserve Bank of St Louis GARCH(1,1) is a two-equation model: Mean Equation: Models the average return (usually close to unpredictable) Variance Equation: Models how risk/volatility changes over time Why the Volatility Part is More Important In financial markets, predicting returns is nearly impossible day-to-day (efficient market hypothesis), but predicting risk is both possible and valuable. Here's why the variance equation matters more: Returns are Random: Daily stock moves are largely unpredictable—yesterday's gain tells you nothing about today's direction. Risk is Predictable: Volatility follows clear patterns—turbulent periods cluster together, and calm follows storms. *Two weeks ago, I had posted a model on Macro-Financial Time Series data: U.S. Treasury Yields & Inflation Dynamics (January 2021- June 2025)