In the last five modules you studied each factor on its own. Now comes the insight that makes all of it actionable: factors are most powerful when combined.
No single factor works every year. Momentum crashes in sharp reversals. Value underperforms for years at a time. Quality lags in bull markets. Low Volatility underperforms when risk appetite is high. But these failure modes don't all happen at the same time, and that's the key. When factors are negatively or lowly correlated, combining them produces a portfolio that's more stable, with lower drawdowns, than any individual factor alone.
This isn't financial magic. It's the mathematics of diversification applied to factor returns rather than individual securities.
Why factor correlation is the key
Consider two investments, each earning 15% annually but with different behaviour:
- Investment A does well in years when B does poorly, and vice versa (negative correlation).
- Investment A and B combined 50/50 earn roughly the same 15% CAGR, but with dramatically lower volatility and drawdowns than either alone.
This is the core principle of factor combination. The returns are additive; the risks are less than additive if the factors are not perfectly correlated. The lower the correlation, the more diversification benefit you get.
Factor correlation matrix
Here is an illustrative correlation matrix for the five factor premiums, meaning the return of each factor over the market, of the kind reported in academic research. Treat the signs as the lesson, not the decimals.
| Factor | Mom | Value | Quality | Low vol | Size |
|---|---|---|---|---|---|
| Momentum | 1.00 | -0.35 | 0.10 | -0.28 | 0.22 |
| Value | -0.35 | 1.00 | -0.20 | 0.05 | 0.18 |
| Quality | 0.10 | -0.20 | 1.00 | 0.32 | -0.15 |
| Low vol | -0.28 | 0.05 | 0.32 | 1.00 | -0.20 |
| Size | 0.22 | 0.18 | -0.15 | -0.20 | 1.00 |
Illustrative correlations of factor premiums, based on the direction of findings in academic literature. Not measured on a specific Indian data set.
The key observations: Momentum and Value are negatively correlated (−0.35), a powerful pairing. Quality and Low Volatility are positively correlated (+0.32), they tend to move together, so combining them adds less diversification. Momentum and Low Volatility are negatively correlated (−0.28), excellent for smoothing the momentum crash problem.
These numbers are illustrative; real correlations vary by period, market regime and how each factor is built. The directions are well documented and more stable than the exact values.
One trap: these are correlations of factor premiums. The long only factor indices you can actually buy all hold the market too, so their raw returns move together. Over eight quarters to October 2026, the quarterly returns of NSE's momentum and value indices had a correlation of about 0.78. The diversification lives in the part of the return that differs, and that part is exactly what saved a blend in 2025, as the worked example below shows.
The two approaches to combining factors
Approach 1: Composite Score (bottom-up)
Calculate a single composite score for each stock by combining multiple factor signals. For example: Composite = 0.4 × Momentum Score + 0.3 × Quality Score + 0.3 × Value Score. Select the top N stocks by composite score.
- Advantage: Finds stocks that rank well on multiple factors simultaneously, strong confirmation signal
- Disadvantage: If one factor dominates (very high momentum stock with poor quality), a stock could rank high overall but is essentially a single-factor bet
Approach 2: Portfolio Blending (top-down)
Build separate factor portfolios and allocate capital between them. E.g., 50% in a momentum portfolio + 50% in a quality portfolio, rebalanced annually.
- Advantage: Clean factor exposure, easier to understand and monitor. Drawdowns from one factor are explicitly offset by the other.
- Disadvantage: Higher turnover and costs. May hold some stocks in both portfolios with offsetting positions.
Research consensus: The composite score approach, combining signals before selection, has tended to do slightly better in studies, because it picks stocks that score well on several factors at once instead of averaging single factor portfolios. NSE's multi factor indices, such as the Nifty Alpha Quality Low-Volatility 30, are built this way.
- Score each factorMomentum, value, quality, each on the same scale.
- CombineAverage the scores, or blend separate portfolios.
- RankOne list from best to worst on the combined view.
- BufferOnly trade when a rank moves meaningfully.
- RebalanceOn a calendar, not after a bad quarter.
Proven factor combinations in Indian markets
What happens to costs in multi-factor strategies
Multi-factor strategies typically have higher turnover than single-factor strategies, because more signals can trigger rebalancing. If momentum says "buy Stock A" but quality says "sell it," the portfolio is in constant tension. This creates more trades, more costs, and more tax events.
The cost-management discipline for multi-factor strategies:
- Use the composite score approach, fewer stocks enter and exit per rebalance compared to blending multiple separate portfolios
- Set rebalancing thresholds, only trade a position if its rank has changed significantly (e.g., a stock drops from top-30 to top-35 doesn't need to be sold)
- Favour quarterly rebalancing for slow-moving factors (value, quality) even if momentum rebalances monthly
Indian factor indices, what NSE actually publishes
Every theoretical factor combination has a live counterpart on NSE Indices. The methodologies are public, the constituents update on a fixed schedule, and the historical data goes back a decade or more. Six worth knowing.
Nifty Alpha 50. Pure momentum. Picks the 50 stocks with the highest alpha, the return above what their beta to the market explains, over the past year. Reviewed quarterly. Since its 2003 base date it has compounded at 20.2% a year on total return, back calculated years included, with five year volatility of 21.7%, the highest of the factor indices here (NSE factsheet, 30 Sep 2026). The textbook momentum factor in Indian form.
Nifty 200 Quality 30. Pure quality. Top 30 names from Nifty 200 ranked on ROE, debt-to-equity stability and EPS growth stability. Semi-annual rebalance. Lower long run return than Alpha 50 and a five year beta of 0.82. Usually the defensive factor, though in the two years to October 2026 it fell more than the Nifty 50 from a high valuation.
Nifty Low Volatility 30. Pure low-vol. The 30 lowest-volatility stocks in Nifty 100, weighted inversely to their volatility. Quarterly rebalance. The calmest of the factor indices: 12.0% volatility in the two years to October 2026. Lags in narrow bull rallies, tends to hold up in stress.
Nifty Alpha Low-Volatility 30. A two-factor blend. Alpha and low volatility combined in one score. Since 2005 it has compounded at 17.7% a year on total return with five year volatility of 13.4%, about the Nifty 50's, against 19.4% for the Nifty200 Momentum 30 (NSE factsheets, 30 Sep 2026, back calculated years included). The composite score approach in published form.
Nifty Alpha Quality Low-Volatility 30. Three-factor composite. Alpha, quality and low volatility combined. Designed to be smoother than any single factor. The closest published NSE index to a true multi factor strategy.
Nifty100 Quality 30. Quality on the Nifty 100. Narrower universe than the 200 Quality 30, slightly different constituent mix, similar return character.
The lesson from these indices is consistent. A multi factor composite rarely beats the best single factor in a given year, but it avoids the worst single factor years, which is the point. Over a decade, the smoother ride tends to win because investors stay invested.
The drawdown rotation no one talks about
Most investor literature stops at "factor diversification reduces drawdown". The harder lesson is that factor leadership rotates inside the drawdown itself. In a sharp drawdown, the order in which factors crack matters.
Momentum cracks first. The names that led on the way up tend to be the most overextended on the way down, so a pure-momentum book often hits its trough before the broader market. Quality holds longer because high-ROE balance sheets weather macro shock better. Low-vol holds longest because the constituents are by definition the least cyclical. Value tends to lead the recovery because deep-discount names benefit first from any sentiment turn.
For a multi-factor strategy, the practical implication is that the rebalance immediately AFTER a sharp drawdown is the highest-information moment of the year. Momentum scores will look terrible on names that just sold off; value scores will look strong on the same names. A thoughtful composite weighting captures the rotation. A naive equal-weight composite captures it less efficiently. This is one of the few legitimate cases where factor weightings can be adjusted with the regime, as long as the rule is set in advance and not in the moment.
A worked example: a 50/50 blend through 2025
Take NSE's momentum and value indices and hold them 50/50, putting the weights back to half each quarter. Here is what that blend did against each on its own, from October 2024 to October 2026.
Rebased to 100 in October 2024: momentum ended at 78.5 after a fall of about 30%; value at 100.9; the 50/50 blend at 89.5, with a deepest quarterly fall of about 22%.
Our calculation from NSE index closes on the 9th of each quarter month, price only, blend rebalanced quarterly, before costs. Quarterly points understate intra quarter falls.
Look at what the blend did and did not do. It did not beat value; nothing in this window did. It cut the momentum fall from about 30% to about 22% at quarterly points, and ended 11 points ahead of momentum alone. If you had to pick one factor in October 2024 without knowing what came next, the blend was the decision you would have regretted least. That is the honest case for combining: not more return in every period, but fewer periods you cannot live through.
MythCombining factors means you get the best of each.
FactYou get something in between, with a smoother path. In the two years to October 2026 a momentum and value blend trailed value and beat momentum. That is the trade you sign up for.
Common mistakes when combining factors
Combining factors that are really the same bet. Quality and low volatility often overlap. Check what the combination adds.
Re-weighting after every bad quarter. Moving money to last quarter's winning factor turns a multi factor plan into factor timing. Module 3.7 covers why that usually fails.
Ignoring costs. More signals can mean more trades. Use rank buffers so small changes do not trigger selling.
Expecting it to top the table. A blend is built not to be last. It will rarely be first.
How this shows up in RupeeCase
The current RupeeCase lineup is mostly single factor, momentum, applied to different universes. The diversification shows up in two other ways. First, across asset classes: Allcap Multi Asset adds 10% debt and 10% gold to the momentum sleeve, and the Dynamic Asset Allocator switches parts of the portfolio between equity, debt and gold by regime. Second, across strategies: you can hold more than one, for example a large cap and a small cap strategy, so their bad stretches do not land at the same time. The comparison pages show the model, back tested records side by side so you can see how they differ before combining them.
How many factors should a portfolio combine?
Two or three that are genuinely different is usually enough: for example momentum with value, or momentum with low volatility. Beyond that the extra diversification is small and the rules get harder to follow.
Should the weights be equal?
Equal weights are a robust default. Weighting by risk, giving the calmer factor more, is a common refinement. Weights tuned to the past are a form of overfitting.
Is a multi asset strategy the same as a multi factor one?
No. Multi factor combines different ways of choosing stocks. Multi asset combines stocks with other assets, such as debt and gold. Both diversify, in different ways.