NJ AMC Investment Process: Best Practices For Championing Rule-Based Investing

• Date : February 06,2025

Investing public money is a responsibility that demands transparency, discipline, and a well-defined process. A sound investment process ensures that every decision is backed by data, research, and tested methodologies rather than emotions or market noise.

At NJ Asset Management Company (NJ AMC), we uphold these principles through a comprehensive, systematic rule-based investment research process that helps avoid human bias and navigate financial markets effectively. Let’s dive into the best practices that make our investment process robust and help us increase the efficacy of our rule-based portfolios.

1. Data Validation, Verification & Cleansing

We treat data as a precious asset, ensuring its accuracy is non-negotiable. To begin with, our Investment Research Analysts collaborate closely with the Data & Analytics Team to validate, verify, and cleanse the data. This step eliminates errors, inconsistencies, and anomalies, ensuring that only clean, comparable, and reliable data guides our decision-making process.

For us, the sanctity of the raw data is as important as the sanctity of defined rules when implementing rule-based investing strategies.

2. Development of Factor Parameters & Hygiene Check

With clean data in hand, we now zoom in on the core building blocks—factor parameters.

Factor investing is like a puzzle—without the right parameters, the picture remains incomplete. Various parameters, such as Return on Capital Employed (ROCE), Return on Equity (ROE), and Free Cash Flow (FCF), among others, act as lenses through which we evaluate potential factors. Each parameter undergoes a rigorous development process, using customised parameter definitions, followed by a hygiene check to guarantee accuracy.

Customising parameter definitions and conditions, instead of using readily available parameter values, is essential for effective factor investing since even small adjustments to parameter definitions can have a significant impact on the backtesting results. For instance, a simple parameter like Return on Equity (ROE) can be computed using various definitions and adjustments for net income.

Once verified, the parameter is added to our Parameter Library, ready to be integrated for future analyses.

3. Parameter Robustness Testing

It's time to separate signal from noise—by subjecting our parameters to rigorous stress tests, we identify the most robust and reliable ones. This step filters out the weak, ensuring only the strongest parameters make the cut. Through comprehensive backtesting across various conditions, we gain valuable insights into their real-world performance.

The structured robustness is carried out by ranking a stock universe based on the chosen parameter, dividing it into different equally sized slices (e.g., terciles, quartiles, quintiles), and analyzing the performance of these slices over various timeframes. Ensuring that an equal number of stocks are distributed across slices is imperative.

The rationale behind this approach is simple: if a parameter effectively contributes to performance, the highest-ranked group should generate better returns than the group below it, which in turn should outperform the next one, and so forth. Additionally, volatility should increase as one moves towards the lower-ranked slices.

For example, take a universe of the 150 largest companies by market capitalization. If these stocks are categorized into three equal groups of 50 each based on a specific parameter (let’s call it ABC), the top slice (highest 50 stocks) would contain the best-ranked stocks, while the lowest slice (bottom 50 stocks) would include the least favourable ones, based on the required characteristics for the parameter ABC.

Parameter name replaced with ABC for illustrative purposes. Past performance may or may not sustain in future.

This example highlights the robustness of parameter ABC, as stocks with higher rankings based on this parameter exhibit a stronger risk-return profile compared to those ranked lower. In other words, the effectiveness of parameter ABC in stock selection is well justified, as incorporating higher-ranked stocks, based on the parameter ABC, into a portfolio can potentially lead to superior returns, reduced volatility, and minimized drawdowns.

Robustness testing plays a crucial role in identifying strong parameters that contribute positively to investment decisions. By systematically filtering out weaker indicators, investors can focus on parameters that enhance portfolio returns while managing risk effectively.

4. Idea Generation & Portfolio Construction

With battle-tested parameters in place, we move to the next frontier—turning insights into unique portfolio strategies. Our research team customises these strategies by defining the universe of stocks, portfolio size, selection criteria, weighting methodology, and the rebalancing frequency and period. This flexibility ensures that portfolios align with the investment objectives.

1. Universe Selection and Portfolio Size: The first step in the investment process is selecting an appropriate universe, which generally consists of all the constituents of a broad-based index viz. Nifty 500. Along with defining a suitable universe, determining the right portfolio size is equally crucial. The portfolio should include an optimal number of stocks to strike a balance, ensuring it is neither too concentrated nor overly diversified, while aligning with the scheme’s objectives.
    
2. Stock Selection: Next, we define the factors that will guide the portfolio generation, along with specific parameters within these factors. Finally, the weight assigned to each parameter and factor within the portfolio is determined.
    
3. Portfolio Allocation: At this stage, we decide how to allocate weights to the individual stocks within the portfolio. This could involve various methodologies, such as equal weighting, market capitalization weighting, factor weighting, and inverse market capitalization weighting among others.
    
4. Portfolio Rebalancing: Rebalancing frequency is established, whether on a monthly, quarterly, half-yearly, or yearly basis.

Once all these elements are defined, the final portfolio is constructed. Thousands of such portfolio strategies are backtested by the Research Team using the proprietary NJ Smart Beta Research Platform.

5. Analysis of The Portfolio Strategy

A portfolio without analysis is like a ship without a compass. It is critical to dissect a portfolio’s performance to ensure it stays on course. We leverage the NJ Smart Beta Research Platform to do so. This process encompasses multiple key aspects:

A. Performance Analysis: We assess various performance metrics to gauge the portfolio’s effectiveness in long duration as well as in different cycles. A few of these metrics include:

1. Point-to-Point Returns: Long-term point-to-point CAGR generated over time.
2. Sharpe Ratio & Sortino Ratio: Risk-adjusted return measures.
3. Rolling Returns: Performance consistency over different holding periods viz. 1 year, 3, years, 5 years and 10 years.
4. Maximum Drawdown: Maximum decline from peak to trough in different market cycles.
5. Calendar Year Returns: Annualized returns for individual calendar years.
6. Volatility: Degree of fluctuations in returns.

This comprehensive analysis sheds light on how our model portfolios stack up against industry standards and benchmark indices such as Nifty 500 TRI, and Nifty 50 TRI.

The table below compares two quality-factor strategies created by a researcher using the NJ Smart Beta Platform: 

Source: Internal research, CMIE, NSE. The period for calculation is 30th September 2006 to 31st December 2024. Past performance may or may not be sustained in future and is not an indication of future return.

Portfolio A outperforms both Nifty 500 TRI and Nifty 500 Quality 50 TRI, delivering higher returns, better risk-adjusted efficiency, and greater resilience during market downturns against Nifty 500 TRI. Portfolio B, while slightly better than Nifty 500 TRI in terms of returns and downside risk, significantly underperforms Nifty 500 Quality 50 TRI across all the key metrics. This makes Portfolio A a better choice, while Portfolio B struggles against quality-focused investing. 

Note that it is imperative to benchmark the factor based strategies not just against broad market indices but also against relevant passive factor based strategy indices such as Nifty 500 Quality 50 TRI as demonstrated above. 

B. Attribution Report: The portfolio attribution analysis includes key performance parameters such as:

1. Average Weight (%): This represents the average proportion of the portfolio allocated to a particular stock or sector across the backtesting period.
2. Total Return in Portfolio (%): This indicates the overall return percentage generated by a particular stock or sector within the portfolio.
3. Total Return Contribution (%): This measures how much of the portfolio’s total return is contributed by a particular stock or sector. Even if a stock or a sector has a high return, its contribution may be low if its weight in the portfolio is small.
4. Gain-To-Loss Ratio: This measures the ratio of the count of instances where a stock or a sector has generated a positive return in the portfolio from its inclusion date to the subsequent exit date to the count of instances where that same stock or sector has generated a negative return in the portfolio from its inclusion date to the subsequent exit date.

C. Churn Analysis:This is a crucial aspect of portfolio evaluation, as it helps assess the frequency and magnitude of transactions within a portfolio. A high churn rate often indicates frequent and heavy buying and selling of securities, which can lead to significant transaction costs, impacting net returns.

Why is Churn Analysis Important?

1. Transaction Costs: Higher churn means more trades, leading to higher brokerage fees, taxes, and other costs that reduce overall returns.
2. Portfolio Stability: Frequent trades may indicate a lack of consistency in strategy, potentially increasing risk.
3. Return Optimization: The goal is to maximize returns while keeping transaction costs minimal to enhance net profitability.

Let us consider an example as follows,

Even though Portfolio A offers a slightly higher return (20% vs. 19%), its high churn (90%) leads to increased transaction costs. In contrast, Portfolio B, with lower churn (30%), retains more of its return. As a result, the net returns of both portfolios may end up being similar, making Portfolio B a more efficient choice.

Thus, lower churn is generally better as it helps optimize returns by minimizing unnecessary costs.

6. Model Finalisation & Implementation

The final stretch—our strategy has been refined, tested, and optimized. Now, it's time to bring it to life. Here our investment committee reviews and evaluates the research output. Only strategies that pass rigorous audit checks and demonstrate a strong potential for risk-adjusted returns are approved for implementation. These strategies are then seamlessly integrated into both new and existing portfolios.

The investment committee does a thorough liquidity analysis of the strategy before implementation to ensure efficient deployment of funds as per the strategy with minimal impact costs and slippages.

Conclusion

At NJ AMC, we’ve taken the guesswork out of investing. By minimizing human biases and adhering to a rule-based approach, we create portfolios that are aligned with predefined objectives. This reduces the need for constant human intervention, ensuring a consistent, reliable investment experience.

FAQs

1) How does NJ AMC test the robustness of its parameters?

Through rigorous backtesting, NJ AMC evaluates each parameter under various market conditions to ensure its reliability and effectiveness in real-world scenarios.

2) What is a rule-based investment process, and why is it important?

A rule-based investment process eliminates human biases and ensures predefined criteria and data-driven analysis guide decisions. 

3) What are factor parameters, and why are they used?

Factor parameters like ROCE or ROE define factors that help guide portfolio decision-making. These parameters are rigorously developed and tested for accuracy to ensure they are accurate.

Investors are requested to take advice from their financial/ tax advisor before making an investment decision.

MUTUAL FUND INVESTMENTS ARE SUBJECT TO MARKET RISKS, READ ALL SCHEME RELATED DOCUMENTS CAREFULLY.