SSRN Id4687044

Timing the Tide: The Impact of Rebalancing Periods in
Momentum Investing in Indian Equities
Rajan Raju∗†
January 2024
Abstract
The frequency of rebalancing is a critical aspect of momentum portfolios with substantial
implications for returns and risk. This paper, the final instalment of our trilogy of papers,
examines portfolios derived from universe sizes of 200, 500, and 750 stocks, with holdings of 15,
30, and 50, across four weighting schemes over varying rebalancing intervals of 1, 2, 3, and 6
months on returns, risks, factor exposures, and other metrics. The critical finding of the study
is that shorter rebalancing periods capture the effect of academic momentum more effectively.
This insight is significant for its implications on strategy optimisation, particularly in the
vibrant and evolving context of Indian equity markets. The paper contributes to academic and
practical aspects of momentum investing, addressing a literature gap, and offering guidance
for investment managers and DIY investors navigating momentum strategies.
Keywords— Factors, Portfolio Construction, Momentum, Rebalancing, Indian Equity
JEL Classification Codes G00, G11, C15
∗
Email: rajanraju@invespar.com, Phone: +65.62380361
†
The author would like to express sincere gratitude to Mr. Abhinav Mehrotra, Founder Stoic Capital Manage-
ment, and two other investment professionals who wish to remain anonymous for their valuable comments in an
earlier draft of the paper.
Electronic copy available at: https://ssrn.com/abstract=4687044

1 Introduction
Momentum strategies have gained traction among wealth advisors and individual investors in India, a
market with distinct characteristics. Drawing on the seminal work by Jegadeesh and Titman (1993), these
strategies find a home in various investment vehicles, ranging from advisor-managed portfolios to indices
such as the Nifty 200 Momentum 30 and a burgeoning community of “Do-It-Yourself" (DIY investors).
This paper is the last of a trilogy of studies that dissect the key aspects of constructing systematic long-
only momentum portfolios in the Indian equity market. Building on our previous work, which explored
the impact of portfolio and universe size on momentum exposure (Raju, 2023a), and the role of weighting
schemes (Raju, 2023b), we now focus on the role of rebalancing periods in long-only momentum strategies.
The frequency of rebalancing is a critical aspect of momentum portfolios with substantial implications
for returns and risk. Following Jegadeesh and Titman (1993), this paper, and much of the literature, uses
monthly data, and the frequency of rebalancing is measured in months. Jegadeesh and Titman report
that the most profitable long/short strategy is rebalancing every 3 months (with a lookback of 12 months
and a skip of 0.25 months). Fama and French (1996) find significant abnormal returns for a strategy re-
balanced monthly. Although academic momentum rebalances monthly (Carhart, 1997; Agarwalla, Jacob,
and Varma, 2013; Raju, 2022b), the popular Nifty 200 Momentum 30 index is rebalanced semiannually.
There is no formal work on the effect of rebalancing frequency in the Indian context for momentum or
momentum tilt strategies. To the best of our knowledge, this paper is the first to formally explore the
effect of rebalancing frequency on momentum portfolios in Indian equities.
Momentum portfolios are about letting winners compound their returns. Rebalancing potentially
neutralises compounding effects within the portfolio by preventing winners from earning higher weights.
Rebalancing has implications for the risks, returns, and costs of the implementation, and all three should
be considered. Cost implications are the easiest to understand and intuit. The low transaction costs
in India make costs a minor consideration. However, slippage costs, given the top-heavy liquidity mi-
crostructure of the Indian equity market, mean that the choice of universe, weight schema, and size of the
strategy will play a significant role in implementation costs. Beyond that, we will not cover costs in this
paper. The return and risk implications are harder to predict and appreciate given both market noise
and complexity, and these are the main focus of the paper.
January 2024 1
Building on insights from our previous explorations of Indian equity momentum strategies, this article,
the final instalment of our trilogy, looks at the effect of rebalancing frequencies of 1, 2, 3, and 6 months on
returns, risks, factor exposures, and other metrics. We examine this on portfolios of 15, 30, and 50 stocks
drawn from a universe of the top 200, 500, and 750 stocks by market capitalisation. Our methodology
extends the earlier evaluations of portfolio weighting schemes, market weight, equal weight, rank weight,
and momentum tilt weighting to paint a holistic picture of the levers that can lead to ‘craftsmanship
alpha’. Our methodology examines 4x3x3x4 = 144 combinations, providing a holistic perspective on the
various combinations of rebalancing frequency, universe, portfolio holdings, and weight schemes.
The study extends the exploration of momentum strategies in the Indian equity market, focussing
on the impact of rebalancing frequencies. It aims to equip investment managers with a nuanced under-
standing of how rebalancing frequencies can refine the efficacy of momentum strategies in the vibrant and
evolving Indian equity markets. Key findings include the effectiveness of shorter rebalancing periods in
capturing the momentum effect and the nuanced interplay between the rebalancing frequency, universe
size, and weighting schemes that affect portfolio performance. These findings contribute significantly to
understanding momentum investing in India’s distinct market.
The paper comprehensively analyses rebalancing frequencies in Indian momentum strategies, a topic
that has been scarcely explored before. It extends existing knowledge by examining various combinations
of universe sizes, portfolio holdings, and weighting schemes, offering a detailed view of how these factors
interact with rebalancing intervals. It contributes to academic and practical fields by providing empiri-
cally grounded insights into the nuances of implementing momentum strategies in Indian equity markets.
Finally, it fills a gap in the existing literature by tailoring the analysis to the unique aspects of the Indian
market.
The rest of this paper is organised as follows. Section 2 discusses our methodology and data sources;
Section 3 presents our results; and we conclude in Section 4.
January 2024 2
2 Methodology and Data
2.1 Portfolio Construction

We follow standard academic practice to calculate the momentum, using the total returns over the
past 12 months but excluding the most recent month. We create portfolios based on three different uni-
verse sizes: the top 200, 500, and 750 companies by market capitalisation, which represent the popular
Nifty200, Nifty500, and Nifty All Cap1 indices. These universes are commonly used in practical mo-
mentum portfolios offered by Asset Managers, wealth advisors through Portfolio Management Services,
and strategies on platforms like Smallcases. We construct four portfolios each month for each universe
containing 15, 30, and 50 holdings. From Raju (2023b), we use four weighting schemes, two based on
market capitalisation (market cap and market cap x Z score) and two not based on market capitalisation
(equal weight and rank weight).
• Market Cap Based Weighting Schemes

Mcapi
– Market Cap: Firms are weighted by their market capitalisation, wi = PN .
j=1 Mcapj
– Market Cap x Z-score: Z-scores from the momentum metrics are computed for all stocks
in the universe for a given period. The z-scores for the portfolio stocks are then normalised.
If a given company’s z-score equals or exceeds 0, its normalised z-score is 1 + z-score. On the
other hand, if its z-score is below 0, then its normalised z-score is (1−z-score) .
1
Companies are
weighted proportionally on this normalised z-score, Z, wi = Z

PN i . Many index providers
j=1 Zj
follow this approach of normalising z-scores for their factor strategy indices.
Z ×Mcapi
Firms are weighted by the product of their z-score and market cap, wi = PN i . Most
j=1 Zj ×Mcapj
momentum factor strategy indices, such as the Nifty 200 Momentum 30 Index2 use this factor
tilt weighting.
• Non-market-cap-based Weighting Schemes
– Equal Weighting: Each firm has the same weight, wi = N.

1
– Rank Weighting: The companies are classified from 1 to N based on their momentum score3
and then weighted proportionally, wi = PN

Ranki
.
j=1 Rankj
Finally, we rebalance portfolios every 1, 2, 3, and 6 months. Standard academic momentum rebal-
ancing is 1 month, while several commercial momentum indices use 6 months. To accurately assess the
1
Approximately 750 stocks.
2
See https://www.niftyindices.com/Methodology/Method_NIFTY_Equity_Indices.pdf.
3
Lowest score is 1 and highest score is N.
January 2024 3
impact of various rebalancing frequencies on momentum portfolios, our methodology involves the creation
of overlapping portfolios with staggered initiation points. For example, with the rebalancing frequency
at 2, we calculate returns for two implementations, one starting at month 0 and the second starting at
month 1. We use a geometric average to compute the time series of monthly returns. Similarly, for a
rebalancing frequency of 6 months, we compute returns for 6 series, starting at month = 0,1,...,5 and
taking the geometric average for the final monthly return time series.
The overlapping nature of the portfolios is central to our approach, as it provides insight into the
performance dynamics under various rebalancing scenarios. This setup mimics real-world investment
scenarios, where portfolios do not all rebalance simultaneously, thus capturing a more realistic range of
outcomes and risk-return profiles. Including multiple starting points for each rebalancing interval ensures
that our analysis is not biased by particular market conditions prevalent at a single time.
Through this method, we create 4 (rebalancing frequency) x 3 (universe) x 3 (portfolio holdings) x 4
(weighting schemes) = 144 portfolios every month and calculate monthly returns for each between March
2005 and August 2023 (222 consecutive months). Even though our data is from September 2004 till
December 2023, we discard 6 months at the two ends as these will have some portfolios with null returns
where the rebalance frequency is 6 months.
Our analysis focuses on evaluating the average monthly returns of these portfolios, taking into account
the distinct starting points. This methodology allows us to dissect the effects of rebalancing frequency
on returns, risk, and factor exposures, thus offering a nuanced understanding of the levers influencing
craftsmanship alpha in the Indian equity market context.
2.2 Summary Statistics

We compute various summary statistics, including mean return, standard deviation, skewness, kurto-
sis, Sharpe ratio, Sortino ratio, effective holdings, average turnover, and others. Using Memmel (2003)
formula, we test the significance of differences in Sharpe ratios:
sri − srj
z=r (1)
1 1 2
T 2(1 − ρi,j ) + 2 (sri + srj2 − sri srj (1 + ρ2i,j ))
January 2024 4
where sr is the Sharpe ratio of, ρi,j is the correlation between the portfolios i and j and T is the number
of observations.
Effective holdings are calculated using the Herfindahl index:
1
Effective Holdings = PN (2)
2
i=1 wi
The turnover is calculated using changes in the allocation to persistent and new stocks based on the
different weighting schemes. One-sided (buy-side) turnover is calculated by adding the weights for new
entrants and an increase in weights at the end of the month prior to the rebalancing for continuing stocks.
Furthermore, turnover is annualised by summing rolling 12-month turnovers of the monthly portfolios,
12
n month turnovers for portfolios rebalanced in n-months. The turnovers are calculated for every month
and an average for the period is taken.
2.3 Returns-based Analysis

The Fama French 5-factor model (Fama and French, 2015) is an extension of their original 3-factor
model (Fama and French, 1993), which only included size, value, and market factors. The momentum
factor was added to the 3-factor model by Carhart (1997). Raju (2022b) details the construction of
Fama-French factors adapted for Indian markets with a financial year ending in March.
We perform returns-based analysis using a Fama-French 5-factor plus momentum model adapted for
the Indian market:
Ri,t = α + βM F + βSM B SM B5 + βHM L HM L + βRM W RM W + βCM A CM A + βW M L W M L + εi,t (3)
where Ri,t is the excess return for portfolio i in period t, α is the intercept, β is the coefficient for the
market factor (M F ), βSM B is the coefficient on the size factor (SM B5 ), βHM L is the coefficient on the
value factor (HM L), βRM W is the coefficient on the profitability factor (RM W ), βCM A is the coefficient
on the investment factor (CM A), βW M L is the coefficient on the momentum factor (W M L), and εi,t is
the residual error term for period t.
This study has limitations. Our data set is relatively shorter than international data, such as the US
market. This creates a sample bias. Additionally, we do not account for transaction and impact costs,
January 2024 5
although the rise of low-cost brokerages and the gradual deepening of market liquidity mitigate this to
some extent. While we consider turnover by design high for momentum strategies, our methodology of
monthly rebalancing without any turnover optimising strategies show turnovers that are likely higher than
any practical strategy. The market’s top-heavy nature may also skew our results.
Furthermore, our analysis is limited to the standard academic momentum factor. It does not necessar-
ily extend to other factors in the asset pricing literature. These systematic portfolio strategies select and
allocate individual stocks that account for multiple factors and risk-based constraints or actively managed
portfolios.
2.4 Data
We use monthly data from LSEG and Datastream due to their comprehensive coverage of Indian
stocks and their reputation for data accuracy. Momentum metrics were derived from Datastream’s total
monthly returns, specifically the 12-month-skip-current-month momentum, aligned with available factor
data between September 2004 and December 2023. The dataset covers 4,907 companies listed on the NSE
or BSE at present or in the past4 . Invespar’s “Data Library: Fama French 3 and 5 Factors and Momentum
Factor for the Indian Market” 5 (Raju, 2022b,a) provides factor data and risk-free rates. Here, the market
factor signifies the value-weighted returns of all pertinent stocks within this period.
3 Results and Discussion

We turn to the empirical results of these diversified portfolio strategies in different universe sizes.
By dissecting the data in multiple dimensions, we aim to build a comprehensive understanding of the
performance characteristics of various levers in implementing a momentum strategy in India.
3.1 Summary Statistics

The analysis of summary statistics for momentum portfolios in different universe sizes, holdings, and
weighting schemes, complemented by varying rebalancing frequencies, shown in Table A1 provides insight
into their performance dynamics in the Indian equity market. To interpret Table A1, readers should
focus on comparing different rows and columns, which represent various combinations of universe size,
rebalancing periods, and weighting schemes. Each cell provides specific statistical metrics such as mean
4
For further universe details, see Raju (2022b).
5
Accessible at https://invespar.com/research.
January 2024 6
annual return, annual standard deviation, skewness, and kurtosis, offering a comprehensive view of the
portfolio performance under different scenarios.
1. Rebalancing Frequency and Portfolio Performance: A key observation is the influence of the
frequency of rebalancing on both absolute and risk-adjusted returns. Consistent with the estab-
lished momentum literature (Jegadeesh and Titman, 1993; Raju and Chandrasekaran, 2019), our
findings suggest that shorter rebalancing periods (e.g., monthly) tend to capture the momentum
effect more efficiently than longer periods, reinforcing the notion that momentum, as a factor, may
dissipate over extended periods. For instance, portfolios with 15 holdings rebalanced monthly in
the top 200 universe exhibit higher mean annual returns (e.g. 19.71% for market weight) com-
pared to those rebalanced semiannually (6 months), with mean annual returns dropping to 17.28%.
This trend is consistently observed across different universe sizes, portfolio holdings, and weighting
schemes, underscoring the temporal nature of momentum.
2. Impact of Universe Size: The choice of universe size significantly impacts the portfolios’ risk-
return profiles. Portfolios constructed from larger universes (500 and 750 stocks) exhibit higher
variability in returns, a trend that is expected given the higher volatility associated with smaller
capitalisation stocks.
The difference in performance due to rebalancing periods is not as pronounced in top 200 universe as
it is in broader 500 and 750 stock universes. Managers could interpret this as momentum portfolios
formed from an universe of top 200 stocks have less variability between each other. Put another
way, they may feel that there is a lower opportunity for craftmanship alpha. We would disagree.
Table A1 shows a spread of more than 10% pa in the mean annual returns for the portfolios in our
sample.
When the number of portfolio holdings and the weighting schema are held constant, the impact
of universe size on the variability of returns from longer rebalancing periods becomes evident. For
example, considering the Rank Weight scheme with 30 holdings, the mean annual return for the
universe of the top 200 stocks with a quarterly rebalance (3 months) period is 22.55%, accompanied
by an annual standard deviation of 28.16%. In contrast, for the same scheme, holdings and rebal-
ancing period with a universe of the top 750 stocks, the mean annual return increases to 31.76%,
but with a significantly higher annual standard deviation of 35.55%. This pattern indicates that
broader universes, while potentially offering higher returns, also entail greater risk, as reflected in
January 2024 7
the higher standard deviations.
This observation aligns with our prior research, which highlighted the trade-offs in selecting universe
size for momentum strategies.
3. Idiosyncratic Risk and Holdings Number: The analysis reveals a direct correlation between
the number of holdings and the idiosyncratic risk, regardless of the frequency of rebalancing. Port-
folios with fewer holdings (15 stocks) exhibit higher standard deviations and varying skewness and
kurtosis, indicative of higher exposure to individual stock risks. For instance, a 50-stock equal-
weight portfolio rebalanced semiannually from the top 500 universe has a mean annual standard
deviation of 29.40% compared to 33.18% for a 15-stock portfolio. Higher returns do not accompany
the higher variability. This finding is crucial for investors and fund managers in balancing diversi-
fication and concentrated exposure to capture momentum.
The relationship between annualised standard deviation with the universe size and number of stocks
is aligned to portfolio theory in genral. However, the 15 stock portfolio weighted by market cap
times z-score drawn from the top 750 universe has lower standard deviation than its 30 stock
counterpart.
4. Influence of Weighting Schemes: Different weighting schemes yield distinct return characteris-
tics. Market Cap and Market Cap x Z-score weighted portfolios demonstrate divergent performance
from Equal Weight and Rank Weight portfolios. This behaviour remains consistent across the re-
balancing frequencies we test.
3.2 Sharpe and Sortino Ratios Analysis

In this subsection, we look at the risk-adjusted performance metrics of our momentum portfolios,
focussing on the Sharpe and Sortino ratios presented in Table A2. These ratios are pivotal in understanding
how the risk-return trade-off varies across different portfolio strategies.
1. Sharpe Ratio Trends Across Rebalancing Frequencies: The data reveals a consistent pattern
where Sharpe ratios generally decrease with longer rebalancing periods. For example, in the top
200 universe with a 1-month rebalancing period, portfolios exhibit higher Sharpe ratios (e.g., a
market weight portfolio shows a Sharpe ratio of 0.40) compared to a 6-month rebalancing period
(where the same strategy shows a Sharpe Ratio of 0.30). This observation suggests that portfolios
January 2024 8
that are rebalanced more frequently are likely to offer better risk-adjusted returns, capturing the
momentum effect more robustly.
2. Sortino Ratio and Rebalancing Frequency: A similar trend is observed with the Sortino
Ratios, which also tend to decrease as the rebalancing periods lengthen. This indicates an increase
in downside risk relative to the risk-free rate in portfolios with less frequent rebalancing, highlighting
the importance of periodic portfolio adjustments in momentum investing.
3. Comparative Analysis Across Universes: When other variables, such as portfolio holdings and
weighting schemes, are held constant, the Sharpe and Sortino ratios vary across different stock uni-
verses. Generally, portfolios within larger universes (e.g., the top 750 stocks) display higher ratios
for shorter rebalancing periods, but this advantage diminishes with the extension of the rebalancing
interval.
4. Impact of Portfolio Holdings and Weighting Schemes: The number of holdings and the
choice of weighting scheme also significantly influence these ratios. Notably, portfolios with fewer
holdings and those employing non-market-cap-based weighting schemes often exhibit higher Sharpe
and Sortino Ratios, indicating more favourable risk-adjusted performance than their counterparts
with larger holdings.
3.3 Drawdown Measures Analysis

This subsection focuses on the drawdown characteristics of the momentum portfolios, as described in
Table A3. Drawdown measures are critical to understanding the risk profiles of investment strategies,
particularly in the context of momentum investing.
1. Impact of Rebalancing Frequency on Maximum Drawdown: The data suggest a notable
trend where maximum drawdowns tend to increase with longer rebalancing intervals. For instance,
within the top 200 universe, portfolios rebalanced monthly show a maximum drawdown of 69.29%
for market weight, which escalates to 73.37% for the same strategy with a 6-month rebalancing
period. This increase in drawdowns for portfolios with less frequent rebalancing indicates a greater
susceptibility to larger losses from momentum crashes, likely due to delayed adjustment to market
movements.
Additionally, having a larger number of holdings in a portfolio reduces drawdowns, as does restrict-
ing the universe to the top 200.
January 2024 9
2. Drawdowns Excluding Global Financial Crisis (GFC): When the GFC period is excluded,
shorter rebalancing periods continue to demonstrate lower maximum drawdowns, signifying better
resilience during market volatility. For example, the maximum drawdown for a market weight
portfolio in the top 200 universe decreases from 69.29% to 42.50% when excluding the GFC, with
a 1-month rebalancing period.
3. Recovery Time from Drawdowns: The duration of maximum drawdowns, particularly when
excluding the GFC, provides insight into the recovery capability of different strategies. Shorter
rebalancing periods generally correspond to quicker recovery times, underlining the effectiveness of
frequent rebalancing in risk management.
The duration of maximum drawdowns does not show a relationship with the rebalancing frequency.
Portfolios drawn from universes of large-cap stocks show shorter drawdowns relative to those from
the top 500 and 750 stocks.
4. Comparison Across Different Universes: Keeping portfolio holdings and weighting schemes
constant, the drawdown profiles differ across various stock universes. Larger universes, such as the
top 750 stocks, exhibit larger drawdowns, especially for portfolios with longer rebalancing periods,
reflecting the amplified risk associated with a broader market exposure.
5. Role of Portfolio Holdings and Weighting Schemes: The extent of drawdowns is also influ-
enced by the number of holdings and the selected weighting schemes. Portfolios with fewer holdings
and those employing non market-cap-based weighting schemes generally experience more significant
drawdowns, indicating a higher volatility and risk profile.
3.4 Additional Portfolio Characteristics Analysis

This subsection focuses on how key portfolio characteristics evolve over different rebalancing frequen-
cies, as presented in Table A4. These characteristics include maximum and minimum weights, effective
holdings, weighted average market capitalisation, P/B ratio, and one-sided turnover.
1. Maximum and Minimum Weight Variations: As the frequency of rebalancing increases, the
maximum weights of the portfolios tend to show a slight increase. For instance, in the top 200
universe, the maximum weight for a market weight portfolio increases from 26.16% in a 1-month
rebalancing to 26.45% in a 6-month rebalancing. This trend might suggest a gradual concentration
in the top holdings as the rebalancing period lengthens. This implies that a longer rebalance period
allows winners to compound their returns.
January 2024 10
In contrast, the minimum weight shows a marginal decrease, reflecting a more pronounced disparity
in the portfolio allocation compared to the academic baseline. However, the variations are relatively
small.
2. Effective Holdings and Diversification Trends: Effective holdings, indicative of diversification,
show a subtle decrease with longer rebalancing periods. For example, the effective holdings for a
market weight portfolio in the top 200 universe decrease from 8.40 in a 1-month rebalancing to
8.30 in a 6-month rebalancing, suggesting a slight reduction in diversification as the frequency of
rebalancing decreases, potentially increasing the risk of the portfolio.
3. Trends in Weighted Average Market Cap and P/B Ratio: The weighted average market cap
shows an increasing trend with longer rebalancing periods, while the weighted average P/B ratio
remains relatively stable. This increase in market cap suggests a gradual shift towards larger-cap
stocks in portfolios with less frequent rebalancing, which could be due to large-cap stocks’ slower
reaction to market changes compared to smaller stocks.
4. Portfolio Turnover across Rebalancing Frequencies: Portfolio turnover is a crucial metric in
systematic portfolios, particularly in the context of momentum portfolios, which often exhibit high
turnover rates. To manage this turnover, practitioners frequently use rebalancing. This analysis
explores the impact of frequency rebalancing on portfolio turnover using our Indian equity dataset.
As an illustrative example, consider a 30-stock portfolio constructed from the top 200 firms, weighted
by market capitalisation times z-score. When rebalanced monthly, this portfolio shows an average
annual one-sided turnover of 3.31 times. However, extending the rebalancing period to semiannu-
ally reduces the annual turnover to 1.36 times. This observation reveals a monotonically decreasing
relationship between the rebalancing frequency and turnover. In practical terms, practitioners often
choose longer rebalancing periods to achieve a reduction in turnover, aligning with this observed
behaviour.
The number of stocks within the portfolio has a direct bearing on turnover. Increasing the number
of stocks mitigates turnover. For example, a 50-stock portfolio rebalanced quarterly is likely to
experience lower turnover compared to a 15-stock portfolio rebalanced quarterly. The rationale
behind this is that a larger portfolio inherently provides “diversification”, reducing the need for
frequent adjustments.
January 2024 11
However, turnover is not solely dependent on the frequency of rebalancing and the size of the port-
folio. It is also influenced by the selection and allocation of stocks. If a rebalanced portfolio includes
a substantial number of new stocks or common stocks with significantly higher weights, turnover
will increase.
To further explore the selection of stocks, Table A5 presents the median percentage of new en-
trants in portfolios of different sizes (15, 30, and 50 stocks) during rebalancing. In our data, larger
portfolios have a lower median percentage of new entrants, and as the rebalancing period extends,
the median percentage of new stocks increases. This explains why a monthly rebalance may have
considerably higher turnover than a semiannual rebalance, even though the latter has a higher per-
centage of new entrants.
Moreover, the size of the portfolio holdings also plays a role. Smaller portfolios, such as 15-stock
portfolios, exhibit a higher percentage of new entrants between rebalancing, resulting in lower
turnover as the portfolio size increases. Slower rebalancing further amplifies turnover, underscoring
the importance of the rebalancing frequency.
In addition, for continuing stocks, the change in weights from the most recent month to the re-
balancing month is another source of turnover. This contribution is relatively small compared to
the percentage of new entrants. Table A6 summarises the mean contribution to turnover arising
from the increase in the weight of the continuing stocks in the portfolio during rebalancing as a
percentage of total turnover.
The change in weights is calculated from the weights at the end of the last rebalance period. So,
for the semiannual rebalanced portfolio rebalanced at the beginning of July, the weights of the con-
tinuing stocks as of the end of June are compared. For example, for a 15-stock portfolio rebalanced
monthly by market weight, 26.5% of the turnover in Table A4 (4.91) comes from the incremental
change in the weights of the continuing stocks. In general, as the rebalancing periods increase, the
contribution of the increase in the weights of the continuing stocks decreases.
It is worth noting that the choice of the stock universe can influence turnover, with portfolios ex-
posed to less liquid small-cap stocks generally incurring higher turnover.
January 2024 12
In conclusion, portfolio turnover is a multifaceted metric influenced by the frequency of rebalancing,
the size of the portfolio, the selection of stocks and the weight changes in the continuing stocks.
This analysis highlights the importance of understanding these dynamics for effective portfolio
management in the realm of quantitative finance research.
3.5 Fama-French 5-Factor plus Momentum Factor Regression Analysis

This subsection looks at the insights derived from the Fama-French 5-Factor plus Momentum Factor
regressions for our diversified 144 momentum portfolios. The regression analysis, from March 2005 to
August 2023, focuses on alpha (intercept), market factor (M F ), beta, size factor (SM B5 ), and momentum
factor (W M L). Table A7 summarises the regression results in three panels: one for each universe. The
coefficients for other factors are not presented. The analysis is structured across different rebalancing
frequencies, holding universe size, portfolio holdings, and weighting schema constant.
1. Alpha (Intercept): None of our 144 portfolios shows statistical evidence of positive alpha over
the Fama French 6-factor model during our observation period. In particular, portfolios with longer
rebalancing periods tend to show more negative alphas, even though the statistical strength is weak
or minimal. For many DIY investors, this result is likely disappointing. However, the alpha in itself
is meaningless without appreciating the changes in the underlying factors.
2. Beta (Market Factor): Beta coefficients are highly significant across all portfolios, consistently
above 1, suggesting a higher sensitivity to market movements, more so than the benchmark. The
trend remains steady regardless of the frequency of rebalancing, portfolio holdings, or universe size.
For instance, a 30-stock rank-weighted portfolio drawn from the top 750 stocks rebalanced monthly
has a market beta of 1.06, increasing to 1.10 when rebalanced quarterly. The higher market beta
as the rebalancing period increases is a result aligned with the higher volatility observed in Table
A1.
3. SM B5 (Size Factor): The significance of the SMB coefficients varies between portfolios. As
a general observation, on average, as the rebalancing period increases, the size increases (lower
coefficients of SM B5 ). This growth in size is another data point indicating that portfolios tend to
compound winners. However, the SM B5 coefficients are affected more significantly by the universe,
the number of portfolio holdings, and the weight schema adopted compared to the rebalancing
period.
4. W M L (Momentum Factor): The strong statistical significance of WML across all portfolios
January 2024 13
underscores the momentum orientation of the strategies. This consistent significance across varying
rebalancing periods reinforces the portfolios’ commitment to momentum strategies. However, on
average, ceteris paribus, as the rebalancing period increases, W M L coefficients decrease. This result
is aligned with Raju and Chandrasekaran (2019).
5. Adjusted R-Squared: The high values of adjusted R-squared across the board suggest that
the Fama-French 5-Factor plus Momentum model robustly explains the return variations of these
portfolios. The rebalancing frequency does not affect the explanatory power of the asset pricing
model.
The analysis underscores the nuanced interplay between the rebalancing frequency, market risk, size
risk, and momentum exposure in momentum tilt portfolios. Variations in alpha across different rebalanc-
ing frequencies and stock universes highlight the complexity of achieving consistent outperformance. The
high beta values across portfolios indicate a more significant risk-return trade-off than the market. The
varied significance of SMB emphasises the selective impact of size risk on portfolio performance. Fur-
thermore, the consistent importance of WML in portfolios reaffirms the strong momentum tilt of these
strategies. Finally, the high adjusted R-squared values validate the effectiveness of the Fama-French 5-
Factor plus Momentum model in capturing the intricacies of portfolio returns in the Indian equity market.
In concluding this section, we reflect on the pivotal role of rebalancing frequency in momentum strategy
portfolios. Our analysis reveals the nuanced impact of rebalancing intervals on portfolio performance
within the Indian equity market. We observe that shorter rebalancing periods capture momentum more
effectively, aligning with the notion that momentum as a factor may dissipate over time. This finding
is consistent across various scenarios, whether considering different universe sizes, holdings, or weighting
schemes. Our study extends the understanding of momentum investing, particularly emphasising the
delicate balance between risk and return in relation to rebalancing frequency. This underscores the
importance of strategic rebalancing in optimising momentum strategy portfolios, a vital consideration for
investors and fund managers who want to harness the full potential of momentum in emerging markets
like India.
3.6 Practical Implications

Using empirical evidence from India, the study has practical implications for investment managers/DIY
investors looking to build their own strategies based on the momentum effect and for investors who want
exposure to the momentum effect. Until now, international studies have been relied on for such purposes,
January 2024 14
and this trilogy of studies offers a very detailed perspective on the Indian experience using a deep and
wide dataset and methodology aligned to the seminal work of Fama-French, thereby minimising sample
or methodology biases.
First, any framework by investment managers and DIY investors to evaluate their proposed momentum-
based strategy should include the following.
1. Rebalancing Frequency Analysis: Managers should consider the impact of various rebalancing
frequencies on portfolio performance. Although shorter intervals have shown efficacy in momen-
tum capture, the choice must be aligned with the strategy’s strategic objectives and risk profile.
Longer rebalancing periods do not automatically mean lower turnover or constant exposure to the
momentum effect.
2. Universe Size and Risk Management: The selection of universe size should be a deliberate
strategy decision balancing the higher return potential of larger universes, and consequently smaller
stocks, against associated risks. Risk management practices should include regular assessments of
market sensitivity and size risk.
3. Diversified Weighting Schemes: Diversification in weighting schemes can potentially enhance
portfolio performance. A mix of market cap-based and non-market cap-based schemes may provide
a balanced approach to capturing momentum while mitigating risks.
Second, a framework for individual investors to evaluate momentum strategies, tailored to their in-
vestment goals and risk tolerance, should include the following.
1. Strategy Selection: Investors should assess momentum strategies by considering the implications
of risks and returns arising from the choice of rebalancing frequency, ensuring alignment with their
investment horizon and risk tolerance.
2. Understanding Factor Exposure: A detailed understanding of the factor exposures of the
chosen strategies is crucial, including considerations of market and size risks in relation to personal
investment goals.
3. Performance Metrics Evaluation: Regular review of key performance metrics such as Sharpe
and Sortino ratios and drawdown analyses is essential to maintain alignment with investment ob-
jectives and risk tolerance.
As we have seen, not all strategies labelled momentum are equal. There is significant variability in
expected outcomes based on implementation choices. Empirical-based anomalies demand regular review
January 2024 15
and assessment.
In our studies, we have not looked at the choice of a momentum metric, the look-back period, and
the skip period for a momentum strategy. This is intentional. First, an existing body of informal work
on these aspects is available on the Internet. While there is a need to formalise this effort using academic
standards, the work is primarily the effort of DIY investors whose search for alpha through ever-increasing
refinements of academic methodology is ongoing. Second, replication studies of Jegadeesh and Titman by
researchers such as Sehgal and Balakrishnan (2002); Ansari and Khan (2012) explore look-back and skip
periods.
4 Conclusion
In this final instalment of our trilogy of papers that examine the implementation choices that a man-
ager looking to exploit the momentum anomaly in Indian equity markets faces, we examine the impact
of rebalancing periods on momentum strategy portfolios. Our comprehensive analysis, spanning differ-
ent universe sizes, portfolio holdings, and weighting schemes, illuminates the intricate dance between
rebalancing frequency and portfolio performance. We find that shorter rebalancing periods are better
at capturing the academic momentum effect, a finding consistent across various market scenarios. This
insight is crucial, highlighting the temporal nature of momentum and its susceptibility to the frequency
of portfolio adjustments.
Our study underscores the importance of strategic rebalancing in optimising momentum strategies.
It offers a granular view that assists investment managers and DIY investors in navigating the complex
landscape of momentum investment. As the Indian market continues to evolve, the insights from this
trilogy will, we hope, serve as a guide for those seeking to harness the full potential of momentum strategies
in India’s vibrant and diverse equity market.
References
Agarwalla, Sobhesh K.; Jacob, Joshy, and Varma, Jayanth R. Four factor model in Indian Equities Market. IIM, Ahmedabad Working
Papers, WP No 2013 09 05, 2013. URL http://www.iimahd.ernet.in/~jrvarma/Indian-Fama-French-Momentum/four-factors-
India-90s-onwards-IIM-WP-Version.pdf.
Ansari, Valeed Ahmad and Khan, Soha. Momentum Anomaly: Evidence from India. Managerial Finance, 38(2):206–223, Jan 2012.
Carhart, Mark M. On Persistence in Mutual Fund Performance. The Journal of Finance, 52(1):57–82, 1997. doi: 10.1111/j.1540-6261.
1997.tb03808.x. URL http://onlinelibrary.wiley.com/doi/abs/10.1111/j.1540-6261.1997.tb03808.x.
January 2024 16
Fama, Eugene F. and French, Kenneth R. Common risk factors in the returns on stocks and bonds. Journal of Financial Economics,
33(1):3–56, 1993. doi: 10.1016/0304-405x(93)90023-5.
Fama, Eugene F and French, Kenneth R. Multifactor Explanations of Asset Pricing Anomalies. The Journal of Finance, 51(1):55–84,
1996.
Fama, Eugene F. and French, Kenneth R. A Five-Factor Asset Pricing Model. Journal of Financial Economics, 116(1):1–22, 2015.
doi: 10.1016/j.jfineco.2014.10. URL https://ideas.repec.org/a/eee/jfinec/v116y2015i1p1-22.html.
Jegadeesh, Narasimhan and Titman, Sheridan. Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency.
The Journal of Finance, 48(1):65, 1993. doi: 10.2307/2328882.
Memmel, Christoph. Performance Hypothesis Testing with the Sharpe Ratio. Finance Letters, 1, 06 2003.
Raju, Rajan. A Five-Factor Asset Pricing Model: Preliminary Evidence from India. Available at SSRN 4190426, page 33, August
2022a. doi: https://dx.doi.org/10.2139/ssrn.4190426. URL https://ssrn.com/abstract=4190426.
Raju, Rajan. Four and Five-Factor Models in the Indian Equities Market. SSRN eLibrary, page 37, March 2022b. doi: http:
//dx.doi.org/10.2139/ssrn.4054146. URL https://ssrn.com/abstract=4054146.
Raju, Rajan. An Examination of Number of Holdings and Universe Size in Momentum Strategies: Evidence from India. SSRN eLibrary,
page 26, May 2023a. doi: 10.2139/ssrn.4453680. URL https://ssrn.com/abstract=4453680.
Raju, Rajan. Watch Your Weight: Navigating Portfolio Weighting Schemes in Indian Momentum Strategies. SSRN eLibrary, page 22,
October 2023b. doi: http://dx.doi.org/10.2139/ssrn.4607933. URL https://ssrn.com/abstract=4607933.
Raju, Rajan and Chandrasekaran, Abhijit. Implementing a Systematic Long-only Momentum Strategy: Evidence From India. SSRN
eJournal, October 2019. URL https://ssrn.com/abstract=3510433o.
Sehgal, Sanjay and Balakrishnan, I. Contrarian and Momentum Strategies in the Indian Capital Market. Vikalpa, 27(1):13–20, January
2002. doi: https://doi.org/10.1177/0256090920020103.
January 2024 17
Table A1: Summary Statistics: Annualised Mean Returns, Annualised Volatility, Skewness and Kurtosis:
March 2005 - August 2023
Holdings 15 30 50
Mcap Mcap Mcap

Scheme Market Equal Rank Market Equal Rank Market Equal Rank
x x x
weight weight weight weight weight weight weight weight weight
ZScore ZScore ZScore
weight weight weight
Rebalancing
Size Period Stats
(months)
Mean Return (Ann %) 19.71 24.91 25.79 20.38 15.74 23.12 23.73 16.22 16.78 21.41 22.49 17.04
Ann Std Dev (%) 30.85 29.29 31.81 35.24 27.15 26.13 27.90 29.48 26.13 24.76 26.06 27.63
1
Skew -0.49 -0.67 -0.52 -0.01 -0.62 -0.86 -0.77 -0.43 0.20 -0.75 -0.84 0.16
Kurtosis 2.48 2.79 1.96 1.94 3.31 3.09 2.79 2.45 8.43 3.40 3.07 6.61
Mean Return (Ann %) 17.25 23.02 24.35 19.08 14.52 22.43 23.07 15.70 15.82 21.56 22.21 16.36
Ann Std Dev (%) 31.06 29.31 32.11 35.19 27.15 26.12 28.12 29.55 25.16 24.72 26.18 26.79
2
Skew -0.59 -0.59 -0.47 -0.09 -0.69 -0.90 -0.74 -0.47 -0.53 -0.88 -0.86 -0.43
Kurtosis 3.07 3.05 2.15 2.25 3.62 3.32 2.95 2.66 4.82 3.25 3.22 3.87
200
Mean Return (Ann %) 17.69 22.93 23.53 19.45 15.05 21.92 22.55 16.32 15.75 21.19 21.88 16.48
Ann Std Dev (%) 31.51 29.25 32.16 35.01 27.37 26.08 28.16 29.66 24.94 24.70 26.19 26.56
3
Skew -0.52 -0.53 -0.36 -0.14 -0.68 -0.85 -0.64 -0.45 -0.75 -0.89 -0.80 -0.59
Kurtosis 2.90 2.57 1.97 2.10 3.65 3.15 2.78 2.68 4.39 3.24 3.08 3.42
Mean Return (Ann %) 17.28 21.60 22.45 17.81 15.82 20.92 21.57 16.63 16.36 20.37 21.05 17.03
Ann Std Dev (%) 33.83 30.18 34.04 36.43 27.93 26.37 29.11 30.25 24.93 24.79 26.67 26.76
6
Skew 0.06 -0.10 0.21 0.10 -0.59 -0.68 -0.25 -0.35 -0.74 -0.80 -0.57 -0.52
Kurtosis 4.91 3.78 3.54 2.99 3.89 3.59 3.63 2.77 4.28 3.54 3.60 3.29
Mean Return (Ann %) 25.65 33.05 33.41 22.40 26.12 33.12 34.31 24.71 20.52 29.75 32.67 21.46
Ann Std Dev (%) 38.02 33.14 35.38 41.32 34.51 31.84 32.63 37.45 29.53 29.16 30.78 32.32
1
Skew -0.55 -0.52 -0.22 -0.15 -0.52 -0.72 -0.59 -0.25 -0.50 -0.75 -0.71 -0.26
Kurtosis 1.92 0.69 0.38 1.54 1.84 1.79 0.97 1.33 2.11 1.86 1.48 1.52
Mean Return (Ann %) 25.74 30.48 29.94 20.98 23.50 30.74 31.48 21.90 20.85 28.74 30.71 21.08
Ann Std Dev (%) 37.67 33.75 35.52 40.74 33.87 31.43 32.62 36.61 29.55 29.19 30.75 32.00
2
Skew -0.50 -0.53 -0.29 -0.11 -0.53 -0.72 -0.60 -0.24 -0.54 -0.73 -0.69 -0.36
Kurtosis 2.03 1.17 0.73 1.64 2.11 1.88 1.29 1.56 2.44 1.97 1.69 1.63
500
Mean Return (Ann %) 22.96 27.95 27.26 19.13 23.34 30.14 30.03 21.74 20.52 27.75 29.48 20.92
Ann Std Dev (%) 37.84 33.50 35.00 40.15 33.98 31.29 32.50 36.05 29.85 29.16 30.71 31.97
3
Skew -0.42 -0.53 -0.30 -0.09 -0.50 -0.64 -0.55 -0.30 -0.56 -0.73 -0.65 -0.39
Kurtosis 2.23 1.27 0.83 1.82 2.28 1.74 1.32 1.66 2.52 1.92 1.65 1.73
Mean Return (Ann %) 18.63 22.97 21.43 14.99 20.58 26.25 25.31 18.21 18.26 24.90 25.51 18.07
Ann Std Dev (%) 38.38 33.18 34.28 39.16 36.72 31.26 32.54 36.91 30.93 29.40 30.88 32.52
6
Skew -0.37 -0.52 -0.33 -0.20 0.18 -0.51 -0.43 -0.08 -0.37 -0.65 -0.52 -0.27
Kurtosis 2.45 1.48 1.07 1.89 4.56 1.96 1.65 2.50 3.02 2.19 1.94 2.14
Mean Return (Ann %) 27.89 30.61 32.23 23.33 26.73 32.04 32.48 24.94 26.23 33.07 33.32 24.93
Ann Std Dev (%) 41.28 34.94 37.59 43.59 37.09 32.67 33.33 39.56 33.32 31.66 32.22 35.78
1
Skew 0.04 -0.50 -0.17 0.05 -0.45 -0.49 -0.50 -0.14 -0.67 -0.68 -0.61 -0.43
Kurtosis 1.28 0.66 0.33 0.80 1.92 1.58 0.69 1.58 1.91 1.82 1.34 1.31
Mean Return (Ann %) 27.03 29.32 29.20 20.07 25.17 29.72 29.76 22.64 23.91 30.95 30.97 22.77
Ann Std Dev (%) 41.30 34.53 36.82 43.25 36.36 32.21 33.07 38.70 32.82 30.89 31.85 35.27
2
Skew 0.17 -0.46 -0.20 0.12 -0.46 -0.66 -0.55 -0.17 -0.70 -0.71 -0.66 -0.42
Kurtosis 1.45 0.98 0.66 0.90 1.78 1.40 1.03 1.43 2.08 1.54 1.39 1.46
750
Mean Return (Ann %) 22.57 27.26 28.15 17.00 23.86 32.93 31.76 21.10 22.94 33.40 33.44 21.71
Ann Std Dev (%) 40.38 33.68 36.01 42.16 36.12 36.20 35.55 37.93 33.34 33.56 34.79 35.14
3
Skew 0.31 -0.43 -0.19 0.23 -0.50 1.15 0.47 -0.27 -0.64 0.35 0.60 -0.44
Kurtosis 1.80 0.87 0.56 1.07 1.97 11.40 5.36 1.43 2.21 5.60 7.04 1.57
Mean Return (Ann %) 14.60 25.78 26.94 10.86 20.14 30.74 28.99 17.02 20.84 30.56 30.86 18.77
Ann Std Dev (%) 39.79 33.52 35.41 40.56 36.95 35.86 33.78 37.33 36.07 33.16 34.03 36.04
6
Skew 0.10 -0.32 -0.11 0.11 -0.20 0.94 -0.22 -0.16 0.06 0.18 0.19 -0.17
Kurtosis 1.33 1.03 0.74 0.79 2.51 9.92 1.56 1.74 4.51 4.50 4.29 2.52
Source: Calculations by the author using LSEG Datastream data and monthly risk-free rate from Raju (2022b),
available at https://www.invespar.com/research
January 2024 18
Table A2: Summary Statistics: Sharpe and Sortino Ratios: March 2005 - August 2023
Holdings 15 30 50
Mcap Mcap Mcap

x x x
Rebalancing
Size Period Stats
(months)
Sharpe Ratio 0.40 0.59 0.57 0.37 0.32 0.60 0.58 0.31 0.37 0.57 0.58 0.36
1 SR t-value vs Mkt Weight 6.75 5.54 -1.52 10.32 8.76 -0.62 7.02 6.63 -0.88
Sortino Ratio 0.80 1.04 1.04 0.82 0.64 1.01 1.00 0.65 0.73 0.96 0.98 0.73
Sharpe Ratio 0.33 0.53 0.52 0.34 0.28 0.58 0.56 0.29 0.35 0.57 0.57 0.35
2 SR t-value vs Mkt Weight 8.05 6.88 0.50 11.23 9.81 0.94 8.85 8.08 -0.20
Sortino Ratio 0.68 0.96 0.98 0.76 0.57 0.97 0.97 0.62 0.66 0.96 0.96 0.68
200
Sharpe Ratio 0.34 0.53 0.50 0.35 0.30 0.56 0.54 0.31 0.35 0.56 0.55 0.36
3 SR t-value vs Mkt Weight 7.81 6.04 0.70 10.60 9.16 1.16 8.91 8.13 0.35
Sortino Ratio 0.70 0.96 0.95 0.77 0.60 0.95 0.95 0.65 0.65 0.94 0.95 0.68
Sharpe Ratio 0.30 0.47 0.44 0.29 0.32 0.52 0.49 0.32 0.37 0.53 0.51 0.37
6 SR t-value vs Mkt Weight 7.25 5.94 -0.56 9.16 7.64 0.05 7.31 6.47 -0.18
Sortino Ratio 0.69 0.91 0.92 0.72 0.64 0.91 0.92 0.67 0.69 0.90 0.91 0.72
Sharpe Ratio 0.47 0.75 0.72 0.36 0.54 0.79 0.80 0.46 0.45 0.75 0.80 0.44
Sortino Ratio 0.95 1.31 1.32 0.84 1.01 1.31 1.36 0.95 0.84 1.24 1.32 0.87
Sharpe Ratio 0.48 0.67 0.62 0.33 0.47 0.73 0.72 0.40 0.46 0.72 0.74 0.43
Sortino Ratio 0.96 1.18 1.16 0.80 0.91 1.23 1.24 0.85 0.85 1.20 1.24 0.85
500
Sharpe Ratio 0.41 0.60 0.56 0.30 0.47 0.71 0.68 0.40 0.44 0.69 0.71 0.43
Sortino Ratio 0.87 1.09 1.06 0.74 0.91 1.21 1.19 0.84 0.83 1.15 1.20 0.84
Sharpe Ratio 0.30 0.47 0.41 0.20 0.36 0.60 0.54 0.30 0.36 0.59 0.58 0.34
Sortino Ratio 0.72 0.89 0.84 0.60 0.82 1.06 1.01 0.72 0.73 1.03 1.03 0.72
Sharpe Ratio 0.49 0.65 0.64 0.36 0.51 0.74 0.73 0.44 0.56 0.79 0.78 0.49
Sortino Ratio 1.05 1.17 1.23 0.88 1.01 1.27 1.28 0.94 1.01 1.31 1.32 0.95
Sharpe Ratio 0.47 0.62 0.58 0.30 0.48 0.68 0.66 0.39 0.50 0.75 0.72 0.43
Sortino Ratio 1.04 1.13 1.13 0.78 0.96 1.17 1.17 0.87 0.92 1.24 1.23 0.88
750
Sharpe Ratio 0.37 0.58 0.57 0.23 0.45 0.69 0.67 0.36 0.46 0.76 0.73 0.41
Sortino Ratio 0.90 1.07 1.11 0.68 0.91 1.35 1.29 0.81 0.88 1.37 1.37 0.84
Sharpe Ratio 0.19 0.54 0.54 0.10 0.35 0.64 0.63 0.27 0.37 0.68 0.67 0.32
Sortino Ratio 0.59 1.03 1.08 0.46 0.78 1.25 1.16 0.67 0.82 1.25 1.25 0.74
January 2024 19
Table A3: Summary Statistics: Drawdown Measures: March 2005 - August 2023
Holdings 15 30 50
Mcap Mcap Mcap

x x x
Rebalancing
Size Period Stats
(months)
Max Drawdown 69.29 72.28 72.25 67.21 67.21 69.91 71.41 66.26 67.87 69.05 70.03 66.90
1 Max Drawdown ex GFC 42.50 34.26 32.36 46.42 32.95 30.42 30.23 36.86 27.56 28.02 29.11 30.89
Max Drawdown Days ex GFC 335 823 335 1,066 335 275 275 335 365 275 275 335
Max Drawdown 71.17 73.03 75.21 69.39 68.45 70.98 72.91 67.71 68.80 69.83 71.50 68.13
Max Drawdown Days ex GFC 335 1,034 335 335 335 275 214 335 365 275 275 335
200
Max Drawdown 72.06 72.93 74.72 70.68 70.39 71.74 73.18 69.76 69.74 70.15 71.73 69.50
Max Drawdown 73.37 73.67 75.60 73.14 72.60 72.69 74.06 72.64 69.77 70.74 72.40 70.52
Max Drawdown 70.35 76.11 74.40 67.84 71.02 76.43 75.67 69.04 70.01 73.41 74.70 68.43
Max Drawdown Days ex GFC 549 670 670 549 335 731 701 335 335 701 701 335
Max Drawdown 72.67 79.52 77.97 71.49 72.54 76.85 77.16 70.91 70.95 73.85 75.35 69.90
500
Max Drawdown 73.86 80.77 79.59 71.72 73.07 76.49 77.60 71.74 71.53 74.32 75.68 70.84
Max Drawdown 74.15 80.64 80.26 73.75 73.74 76.51 78.03 73.77 73.42 75.60 76.57 73.46
Max Drawdown 77.91 78.33 74.37 74.44 71.09 71.37 73.28 68.48 70.77 71.78 71.99 68.76
Max Drawdown Days ex GFC 701 731 823 2,404 335 914 853 549 335 945 914 335
Max Drawdown 75.85 79.14 77.29 77.24 73.07 76.19 77.18 70.64 72.46 73.49 75.32 70.87
750
Max Drawdown 74.39 79.44 77.32 77.45 74.01 76.37 77.48 71.84 72.87 72.20 75.32 71.53
Max Drawdown 75.11 79.19 77.24 73.46 74.47 78.74 78.83 73.61 73.91 74.38 76.98 73.59
Max Drawdown Days ex GFC 549 1,004 882 549 457 1,004 1,004 457 335 945 945 335
January 2024 20
Table A4: Summary Statistics: Additional Portfolio Characteristics: March 2005 - August 2023
Holdings 15 30 50
Mcap Mcap Mcap
x x x
Rebalancing
Size Period Stats
(months)
Max Weight 26.16 6.67 12.50 28.88 20.01 3.33 6.45 20.58 17.31 2.00 3.92 16.51
Min Weight 2.13 6.67 0.83 1.78 0.87 3.33 0.22 0.74 0.43 2.00 0.08 0.37
Eff Holdings 8.40 15.00 11.60 7.50 13.40 30.00 22.90 12.60 16.20 50.00 37.90 16.20
1
Wtd Average Market Cap 794.10 329.00 318.50 785.80 1,090.40 357.30 338.10 1,038.20 1,308.00 385.20 359.50 1,236.10
Wtd Average P/B 13.29 9.52 10.84 15.94 10.38 8.01 9.12 12.68 8.98 7.43 8.12 10.81
Average Turnover 4.61 3.75 3.23 3.74 4.18 3.28 2.75 3.31 3.51 2.78 2.41 2.79
Max Weight 26.22 6.67 12.50 28.95 20.07 3.33 6.45 20.63 17.34 2.00 3.92 16.54
Min Weight 2.13 6.67 0.83 1.77 0.87 3.33 0.22 0.74 0.43 2.00 0.08 0.37
Eff Holdings 8.40 15.00 11.60 7.50 13.40 30.00 22.90 12.60 16.20 50.00 37.90 16.10
2
Wtd Average P/B 13.31 9.52 10.83 15.95 10.39 8.01 9.12 12.69 8.99 7.42 8.11 10.81
200
Max Weight 26.28 6.67 12.50 29.02 20.13 3.33 6.45 20.69 17.36 2.00 3.92 16.57
Min Weight 2.12 6.67 0.83 1.77 0.86 3.33 0.22 0.74 0.43 2.00 0.08 0.37
Eff Holdings 8.40 15.00 11.60 7.40 13.30 30.00 22.90 12.60 16.10 50.00 37.90 16.00
3
Wtd Average P/B 13.34 9.53 10.83 15.99 10.41 8.01 9.12 12.72 9.00 7.42 8.11 10.83
Max Weight 26.45 6.67 12.50 29.24 20.29 3.33 6.45 20.87 17.40 2.00 3.92 16.62
Min Weight 2.10 6.67 0.83 1.75 0.85 3.33 0.22 0.73 0.42 2.00 0.08 0.37
Eff Holdings 8.30 15.00 11.60 7.40 13.20 30.00 22.90 12.60 15.80 50.00 37.90 15.90
6
Wtd Average P/B 13.39 9.51 10.79 16.05 10.44 8.00 9.10 12.76 9.02 7.42 8.10 10.87
Max Weight 31.81 6.67 12.50 32.61 23.18 3.33 6.45 25.08 20.28 2.00 3.92 21.08
Min Weight 1.52 6.67 0.83 1.18 0.62 3.33 0.22 0.48 0.30 2.00 0.08 0.24
Eff Holdings 6.40 15.00 11.60 6.10 9.90 30.00 22.90 9.40 13.30 50.00 37.90 12.80
1
Wtd Average Market Cap 403.90 118.80 117.20 389.20 564.30 128.70 121.90 551.10 715.80 142.10 130.20 690.50
Wtd Average P/B 18.58 11.06 13.13 21.11 14.32 8.56 10.37 17.36 11.85 7.45 8.76 14.74
Max Weight 31.87 6.67 12.50 32.67 23.21 3.33 6.45 25.14 20.33 2.00 3.92 21.13
Min Weight 1.51 6.67 0.83 1.17 0.62 3.33 0.22 0.48 0.30 2.00 0.08 0.24
Eff Holdings 6.40 15.00 11.60 6.10 9.90 30.00 22.90 9.30 13.30 50.00 37.90 12.80
2
Wtd Average P/B 18.57 11.04 13.11 21.11 14.33 8.55 10.35 17.38 11.85 7.43 8.75 14.75
500
Max Weight 31.92 6.67 12.50 32.73 23.27 3.33 6.45 25.19 20.38 2.00 3.92 21.18
Min Weight 1.51 6.67 0.83 1.17 0.62 3.33 0.22 0.48 0.30 2.00 0.08 0.24
Eff Holdings 6.40 15.00 11.60 6.10 9.90 30.00 22.90 9.30 13.30 50.00 37.90 12.70
3
Wtd Average P/B 18.60 11.04 13.12 21.14 14.36 8.55 10.35 17.42 11.87 7.43 8.75 14.78
Max Weight 32.15 6.67 12.50 32.94 23.44 3.33 6.45 25.37 20.51 2.00 3.92 21.35
Min Weight 1.50 6.67 0.83 1.16 0.61 3.33 0.22 0.47 0.29 2.00 0.08 0.23
Eff Holdings 6.30 15.00 11.60 6.10 9.70 30.00 22.90 9.20 13.10 50.00 37.90 12.70
6
Wtd Average P/B 18.62 11.01 13.09 21.19 14.41 8.55 10.34 17.48 11.93 7.43 8.75 14.86
Max Weight 35.80 6.67 12.50 37.65 27.40 3.33 6.45 28.23 21.50 2.00 3.92 23.05
Min Weight 1.24 6.67 0.83 0.83 0.47 3.33 0.22 0.34 0.24 2.00 0.08 0.18
Eff Holdings 6.50 15.00 11.60 5.30 8.90 30.00 22.90 8.50 11.70 50.00 37.90 11.20
1
Wtd Average P/B 20.47 11.60 13.28 22.35 16.84 9.40 11.00 19.50 14.12 8.08 9.44 17.16
Max Weight 35.83 6.67 12.50 37.70 27.46 3.33 6.45 28.28 21.54 2.00 3.92 23.11
Min Weight 1.24 6.67 0.83 0.83 0.47 3.33 0.22 0.34 0.24 2.00 0.08 0.18
Eff Holdings 6.50 15.00 11.60 5.30 8.90 30.00 22.90 8.50 11.70 50.00 37.90 11.20
2
Wtd Average P/B 20.47 11.57 13.24 22.34 16.83 9.37 10.97 19.49 14.13 8.06 9.41 17.17
750
Max Weight 35.86 6.67 12.50 37.76 27.52 3.33 6.45 28.35 21.58 2.00 3.92 23.16
Min Weight 1.24 6.67 0.83 0.82 0.47 3.33 0.22 0.34 0.24 2.00 0.08 0.17
Eff Holdings 6.50 15.00 11.60 5.30 8.80 30.00 22.90 8.50 11.60 50.00 37.90 11.20
3
Wtd Average P/B 20.51 11.55 13.21 22.38 16.86 9.36 10.95 19.53 14.15 8.05 9.40 17.20
Max Weight 36.09 6.67 12.50 37.99 27.74 3.33 6.45 28.54 21.75 2.00 3.92 23.33
Min Weight 1.24 6.67 0.83 0.82 0.46 3.33 0.22 0.33 0.23 2.00 0.08 0.17
Eff Holdings 6.50 15.00 11.60 5.30 8.80 30.00 22.90 8.40 11.50 50.00 37.90 11.00
6
Wtd Average P/B 20.52 11.50 13.15 22.42 16.89 9.33 10.90 19.58 14.19 8.03 9.36 17.25
Note: Effective holdings is a measure of portfolio diversification and is calculated using Equation 2. The weighted
average market cap per holding, in Rs. billion, is the time series average of the monthly market cap of the
portfolio divided by the number of holdings. The weighted average P / B is the time series average of the
portfolio P/B. Turnover is a one-sided turnover.
January 2024 21
Table A5: Median Percentage of New Stocks on Rebalance: March 2005 - August 2023
Holdings 15 30 50
Rebalancing
Size Period
(months)
1 33.3 26.7 22.0
2 46.7 36.7 31.0
200
3 53.3 46.7 38.0
6 73.3 63.3 56.0
1 33.3 30.0 26.0
2 46.7 40.0 36.0
500
3 53.3 50.0 46.0
6 80.0 73.3 68.0
1 33.3 30.0 26.0
2 46.7 41.7 38.0
750
3 53.3 50.0 48.0
6 80.0 73.3 70.0
Note: The table shows the median percentage of stocks in an N-stock portfolio that are new at the point of
rebalancing. As this is a measure in the stock selection phase, it applies regardless of the weight schema adopted,
which is a stock allocation decision.
Table A6: Mean Percentage Contribution to Annual Turnover due to Increase of Weights of Stocks
Remaining in Portfolios Between two Rebalancing Periods: March 2005 - August 2023
Holdings 15 30 50
Mcap Mcap Mcap
x x x
Rebalancing
Size Period
(months)
1 19.6 15.8 31.2 35.1 19.0 12.1 28.6 36.4 22.9 14.3 29.2 28.6
2 16.1 11.1 20.8 33.3 18.5 13.0 19.0 26.1 20.8 10.0 22.2 20.0
200
3 8.3 9.1 15.0 28.6 18.2 5.3 16.7 21.1 16.7 6.2 18.7 17.6
6 13.3 13.3 13.3 26.7 14.3 7.7 14.3 21.4 8.3 8.3 16.7 16.7
1 20.8 21.1 34.4 44.1 24.4 22.2 27.6 39.4 24.4 20.6 25.9 38.2
2 25.8 11.1 25.0 41.7 26.7 16.0 18.2 37.5 24.1 17.4 19.0 29.2
500
3 16.7 26.1 27.3 33.3 17.4 14.3 25.0 40.0 21.7 10.0 16.7 25.0
6 25.0 12.5 20.0 20.0 13.3 13.3 13.3 21.4 20.0 7.1 14.3 14.3
1 26.5 20.5 33.3 50.0 23.4 22.2 34.5 43.7 26.7 20.6 22.2 45.2
2 21.2 7.4 20.8 45.8 20.0 11.5 21.7 39.1 20.0 12.5 14.3 30.4
750
3 24.0 26.1 28.6 45.0 20.8 18.2 23.8 45.0 17.4 14.3 15.8 25.0
6 25.0 12.5 13.3 20.0 13.3 13.3 13.3 21.4 13.3 7.1 7.1 14.3
Note: The table shows the mean percentage contribution to annual turnover shown in Table A4 due to the
increase in the weight of stocks in an N-stock portfolio that continue between two rebalancing periods.
January 2024 22
Table A7: Fama-French 5-Factor + Momentum Factor Regressions: March 2005 - August 2023
Panel A: Top 200

Holdings 15 30 50
Mcap Mcap Mcap
x x x
Rebalancing
Period Stats
(months)
FF6 α -0.49 -0.19 -0.03 -0.39 -0.74*** -0.22 -0.22 -0.68** -0.51*** -0.25** -0.25* -0.53***
α SE (0.33) (0.25) (0.35) (0.44) (0.22) (0.15) (0.21) (0.27) (0.15) (0.11) (0.15) (0.18)
MF β 1.12*** 1.10*** 1.09*** 1.15*** 1.09*** 1.04*** 1.08*** 1.12*** 1.10*** 1.03*** 1.05*** 1.13***
β SE (0.08) (0.06) (0.09) (0.10) (0.04) (0.04) (0.05) (0.05) (0.05) (0.02) (0.03) (0.05)
1 SM B5 0.10 0.31*** 0.29** 0.11 0.08 0.31*** 0.30*** 0.06 -0.10 0.26*** 0.29*** -0.09
SM B5 SE (0.11) (0.10) (0.13) (0.14) (0.08) (0.07) (0.09) (0.09) (0.06) (0.05) (0.06) (0.07)
W ML 0.83*** 0.69*** 0.72*** 0.92*** 0.71*** 0.57*** 0.63*** 0.77*** 0.70*** 0.48*** 0.56*** 0.74***
W M L SE (0.10) (0.08) (0.11) (0.13) (0.06) (0.05) (0.07) (0.07) (0.06) (0.03) (0.04) (0.06)
Adj. R-squared 0.75 0.82 0.74 0.66 0.86 0.90 0.86 0.82 0.93 0.94 0.91 0.91
FF6 α -0.58* -0.25 -0.09 -0.44 -0.83*** -0.25* -0.22 -0.73*** -0.61*** -0.24** -0.26* -0.59***
α SE (0.33) (0.24) (0.34) (0.42) (0.21) (0.15) (0.21) (0.26) (0.14) (0.12) (0.15) (0.18)
MF β 1.13*** 1.11*** 1.11*** 1.17*** 1.10*** 1.05*** 1.09*** 1.14*** 1.07*** 1.04*** 1.06*** 1.11***
β SE (0.07) (0.06) (0.08) (0.08) (0.04) (0.03) (0.05) (0.05) (0.02) (0.02) (0.03) (0.03)
2 SM B5 0.06 0.26*** 0.25** 0.07 0.08 0.29*** 0.27*** 0.06 -0.02 0.27*** 0.28*** -0.02
SM B5 SE (0.10) (0.09) (0.12) (0.13) (0.07) (0.06) (0.08) (0.09) (0.04) (0.05) (0.06) (0.06)
W ML 0.80*** 0.67*** 0.71*** 0.90*** 0.70*** 0.55*** 0.62*** 0.76*** 0.63*** 0.45*** 0.54*** 0.68***
W M L SE (0.10) (0.07) (0.10) (0.13) (0.06) (0.05) (0.06) (0.07) (0.04) (0.03) (0.05) (0.05)
Adj. R-squared 0.78 0.84 0.75 0.69 0.88 0.91 0.87 0.83 0.93 0.94 0.92 0.91
FF6 α -0.48 -0.22 -0.12 -0.36 -0.76*** -0.27* -0.23 -0.67** -0.60*** -0.26** -0.26 -0.58***
α SE (0.34) (0.25) (0.35) (0.42) (0.21) (0.16) (0.22) (0.26) (0.14) (0.12) (0.16) (0.17)
MF β 1.15*** 1.10*** 1.13*** 1.18*** 1.11*** 1.05*** 1.09*** 1.14*** 1.07*** 1.03*** 1.06*** 1.10***
β SE (0.07) (0.06) (0.09) (0.08) (0.04) (0.04) (0.06) (0.05) (0.02) (0.03) (0.04) (0.03)
3 SM B5 0.02 0.25*** 0.22* 0.06 0.07 0.28*** 0.25*** 0.07 -0.00 0.26*** 0.27*** 0.01
SM B5 SE (0.11) (0.09) (0.12) (0.14) (0.07) (0.06) (0.08) (0.10) (0.04) (0.05) (0.06) (0.06)
W ML 0.75*** 0.63*** 0.68*** 0.82*** 0.67*** 0.52*** 0.59*** 0.73*** 0.59*** 0.44*** 0.52*** 0.63***
W M L SE (0.11) (0.09) (0.11) (0.14) (0.07) (0.06) (0.08) (0.09) (0.04) (0.04) (0.06) (0.06)
Adj. R-squared 0.76 0.83 0.76 0.69 0.87 0.90 0.86 0.82 0.93 0.94 0.91 0.91
FF6 α -0.50 -0.32 -0.25 -0.44 -0.65*** -0.33* -0.31 -0.58** -0.49*** -0.31** -0.31* -0.45**
α SE (0.36) (0.25) (0.33) (0.43) (0.22) (0.17) (0.22) (0.29) (0.15) (0.13) (0.17) (0.20)
MF β 1.26*** 1.18*** 1.26*** 1.28*** 1.14*** 1.09*** 1.16*** 1.18*** 1.08*** 1.05*** 1.10*** 1.12***
β SE (0.07) (0.07) (0.09) (0.08) (0.04) (0.04) (0.06) (0.05) (0.03) (0.03) (0.04) (0.04)
6 SM B5 -0.12 0.17* 0.14 -0.04 0.02 0.23*** 0.19** 0.03 -0.01 0.24*** 0.22*** -0.00
SM B5 SE (0.13) (0.11) (0.14) (0.14) (0.07) (0.06) (0.09) (0.09) (0.04) (0.05) (0.07) (0.06)
W ML 0.79*** 0.63*** 0.70*** 0.79*** 0.63*** 0.51*** 0.59*** 0.65*** 0.49*** 0.41*** 0.51*** 0.51***
W M L SE (0.15) (0.11) (0.14) (0.17) (0.09) (0.07) (0.10) (0.12) (0.06) (0.05) (0.07) (0.08)
Adj. R-squared 0.77 0.84 0.77 0.70 0.87 0.91 0.86 0.81 0.92 0.94 0.91 0.88
Panel B: Top 500

Holdings 15 30 50
Mcap Mcap Mcap
x x x
Rebalancing
Period Stats
(months)
FF6 α -0.07 0.07 0.20 -0.30 -0.21 -0.02 0.09 -0.29 -0.52** -0.13 -0.00 -0.45
α SE (0.56) (0.37) (0.45) (0.58) (0.43) (0.25) (0.31) (0.47) (0.25) (0.18) (0.24) (0.32)
MF β 1.16*** 1.07*** 1.05*** 1.20*** 1.17*** 1.12*** 1.11*** 1.21*** 1.16*** 1.10*** 1.11*** 1.19***
β SE (0.13) (0.07) (0.08) (0.12) (0.10) (0.04) (0.05) (0.10) (0.05) (0.03) (0.04) (0.06)
1 SM B5 0.44*** 0.91*** 0.93*** 0.46*** 0.44*** 0.90*** 0.92*** 0.44*** 0.22** 0.78*** 0.86*** 0.25**
SM B5 SE (0.16) (0.10) (0.12) (0.17) (0.14) (0.08) (0.09) (0.14) (0.10) (0.07) (0.07) (0.11)
W ML 0.82*** 0.64*** 0.65*** 0.93*** 0.83*** 0.65*** 0.63*** 0.90*** 0.71*** 0.56*** 0.61*** 0.78***
W M L SE (0.17) (0.09) (0.11) (0.17) (0.14) (0.08) (0.09) (0.14) (0.08) (0.05) (0.07) (0.09)
Adj. R-squared 0.58 0.73 0.63 0.54 0.70 0.86 0.79 0.65 0.82 0.90 0.86 0.76
FF6 α -0.11 -0.15 -0.07 -0.40 -0.31 -0.12 -0.05 -0.42 -0.47* -0.18 -0.09 -0.45
α SE (0.48) (0.33) (0.41) (0.53) (0.39) (0.24) (0.30) (0.44) (0.25) (0.18) (0.23) (0.31)
MF β 1.25*** 1.12*** 1.10*** 1.24*** 1.16*** 1.11*** 1.12*** 1.20*** 1.15*** 1.09*** 1.10*** 1.18***
β SE (0.10) (0.05) (0.06) (0.10) (0.09) (0.04) (0.05) (0.09) (0.05) (0.03) (0.04) (0.06)
2 SM B5 0.37** 0.97*** 0.96*** 0.46** 0.37*** 0.87*** 0.90*** 0.41*** 0.25*** 0.79*** 0.85*** 0.29***
SM B5 SE (0.16) (0.10) (0.11) (0.18) (0.13) (0.08) (0.09) (0.15) (0.09) (0.07) (0.07) (0.11)
W ML 0.79*** 0.64*** 0.66*** 0.84*** 0.79*** 0.63*** 0.63*** 0.84*** 0.68*** 0.56*** 0.60*** 0.74***
W M L SE (0.17) (0.09) (0.10) (0.18) (0.13) (0.07) (0.08) (0.15) (0.07) (0.05) (0.07) (0.09)
Adj. R-squared 0.64 0.78 0.68 0.57 0.73 0.86 0.81 0.67 0.84 0.90 0.87 0.78
FF6 α -0.26 -0.28 -0.24 -0.52 -0.22 -0.11 -0.11 -0.35 -0.44* -0.21 -0.13 -0.42
α SE (0.47) (0.32) (0.38) (0.51) (0.40) (0.25) (0.29) (0.43) (0.26) (0.19) (0.24) (0.31)
MF β 1.26*** 1.12*** 1.10*** 1.24*** 1.17*** 1.11*** 1.12*** 1.20*** 1.15*** 1.09*** 1.10*** 1.17***
β SE (0.09) (0.05) (0.06) (0.10) (0.09) (0.05) (0.05) (0.09) (0.06) (0.04) (0.04) (0.07)
3 SM B5 0.35** 0.93*** 0.93*** 0.45** 0.29** 0.84*** 0.86*** 0.36** 0.25** 0.77*** 0.82*** 0.29**
SM B5 SE (0.17) (0.10) (0.11) (0.18) (0.13) (0.08) (0.09) (0.15) (0.10) (0.07) (0.08) (0.12)
W ML 0.75*** 0.63*** 0.67*** 0.79*** 0.74*** 0.60*** 0.62*** 0.77*** 0.66*** 0.55*** 0.59*** 0.70***
W M L SE (0.17) (0.09) (0.10) (0.19) (0.14) (0.08) (0.08) (0.15) (0.09) (0.06) (0.07) (0.11)
Adj. R-squared 0.64 0.78 0.70 0.57 0.73 0.86 0.81 0.68 0.82 0.90 0.87 0.77
FF6 α -0.41 -0.50* -0.54 -0.64 -0.36 -0.30 -0.35 -0.50 -0.59** -0.36* -0.34 -0.57*
α SE (0.47) (0.30) (0.35) (0.50) (0.40) (0.24) (0.28) (0.43) (0.28) (0.20) (0.23) (0.33)
MF β 1.28*** 1.14*** 1.10*** 1.22*** 1.32*** 1.15*** 1.15*** 1.28*** 1.21*** 1.12*** 1.14*** 1.22***
β SE (0.08) (0.05) (0.06) (0.10) (0.09) (0.05) (0.06) (0.08) (0.05) (0.04) (0.05) (0.06)
6 SM B5 0.27* 0.84*** 0.86*** 0.37** 0.08 0.73*** 0.76*** 0.20 0.16* 0.71*** 0.74*** 0.21*
SM B5 SE (0.16) (0.09) (0.10) (0.17) (0.16) (0.09) (0.10) (0.14) (0.10) (0.08) (0.09) (0.11)
W ML 0.68*** 0.56*** 0.59*** 0.66*** 0.81*** 0.57*** 0.59*** 0.75*** 0.68*** 0.51*** 0.56*** 0.67***
W M L SE (0.19) (0.10) (0.11) (0.21) (0.16) (0.08) (0.09) (0.17) (0.11) (0.07) (0.08) (0.13)
Adj. R-squared 0.66 0.80 0.72 0.60 0.75 0.86 0.82 0.71 0.83 0.89 0.86 0.78
January 2024 23
Fama-French 5-Factor + Momentum Factor Regressions: March 2005 - August 2023...contd
Panel C: Top 750

Holdings 15 30 50
Mcap Mcap Mcap
x x x
Rebalancing
Period Stats
(months)
FF6 α 0.29 0.13 0.31 -0.09 -0.13 -0.02 0.12 -0.22 -0.18 0.00 0.06 -0.25
α SE (0.62) (0.42) (0.50) (0.63) (0.49) (0.31) (0.36) (0.53) (0.39) (0.24) (0.28) (0.44)
MF β 1.17*** 1.02*** 1.05*** 1.20*** 1.24*** 1.09*** 1.06*** 1.24*** 1.19*** 1.11*** 1.09*** 1.23***
β SE (0.16) (0.08) (0.10) (0.14) (0.10) (0.06) (0.07) (0.10) (0.09) (0.05) (0.06) (0.09)
1 SM B5 0.36* 1.03*** 1.04*** 0.54*** 0.44*** 0.97*** 0.99*** 0.49*** 0.42*** 0.98*** 0.99*** 0.44***
SM B5 SE (0.20) (0.11) (0.12) (0.20) (0.14) (0.09) (0.10) (0.16) (0.14) (0.08) (0.09) (0.14)
W ML 0.75*** 0.49*** 0.45*** 0.79*** 0.80*** 0.58*** 0.52*** 0.86*** 0.69*** 0.55*** 0.55*** 0.76***
W M L SE (0.16) (0.12) (0.14) (0.17) (0.15) (0.10) (0.11) (0.16) (0.13) (0.08) (0.10) (0.13)
Adj. R-squared 0.53 0.66 0.56 0.48 0.65 0.79 0.72 0.60 0.72 0.87 0.81 0.67
FF6 α 0.14 -0.02 0.06 -0.35 -0.17 -0.15 -0.07 -0.34 -0.28 -0.09 -0.06 -0.36
α SE (0.55) (0.38) (0.46) (0.58) (0.47) (0.28) (0.34) (0.50) (0.36) (0.23) (0.27) (0.41)
MF β 1.28*** 1.08*** 1.10*** 1.22*** 1.22*** 1.09*** 1.09*** 1.22*** 1.18*** 1.09*** 1.09*** 1.22***
β SE (0.11) (0.07) (0.09) (0.11) (0.10) (0.06) (0.07) (0.10) (0.08) (0.05) (0.06) (0.09)
2 SM B5 0.35* 0.98*** 0.97*** 0.59*** 0.37** 0.96*** 0.97*** 0.47*** 0.38*** 0.95*** 0.96*** 0.43***
SM B5 SE (0.19) (0.11) (0.12) (0.20) (0.14) (0.09) (0.10) (0.16) (0.13) (0.08) (0.09) (0.14)
W ML 0.74*** 0.50*** 0.45*** 0.74*** 0.77*** 0.55*** 0.51*** 0.80*** 0.67*** 0.54*** 0.54*** 0.72***
W M L SE (0.15) (0.11) (0.14) (0.17) (0.15) (0.10) (0.11) (0.17) (0.12) (0.08) (0.09) (0.13)
Adj. R-squared 0.58 0.70 0.60 0.50 0.67 0.82 0.75 0.61 0.75 0.88 0.83 0.69
FF6 α -0.16 -0.16 -0.00 -0.56 -0.19 0.12 0.09 -0.40 -0.29 0.10 0.14 -0.37
α SE (0.50) (0.36) (0.43) (0.55) (0.47) (0.37) (0.38) (0.49) (0.38) (0.31) (0.34) (0.41)
MF β 1.29*** 1.07*** 1.09*** 1.21*** 1.22*** 1.10*** 1.10*** 1.21*** 1.20*** 1.10*** 1.10*** 1.22***
β SE (0.10) (0.06) (0.09) (0.11) (0.10) (0.06) (0.07) (0.10) (0.08) (0.05) (0.06) (0.09)
3 SM B5 0.32* 0.96*** 0.97*** 0.58*** 0.33** 0.92*** 0.93*** 0.44*** 0.32** 0.92*** 0.93*** 0.39***
SM B5 SE (0.18) (0.10) (0.11) (0.19) (0.15) (0.09) (0.10) (0.16) (0.13) (0.08) (0.09) (0.14)
W ML 0.71*** 0.50*** 0.46*** 0.69*** 0.72*** 0.63*** 0.58*** 0.74*** 0.65*** 0.59*** 0.60*** 0.68***
W M L SE (0.15) (0.10) (0.13) (0.18) (0.15) (0.13) (0.13) (0.17) (0.13) (0.11) (0.12) (0.15)
Adj. R-squared 0.62 0.74 0.63 0.51 0.67 0.67 0.67 0.61 0.74 0.76 0.71 0.69
FF6 α -0.59 -0.23 -0.08 -0.87 -0.35 0.00 -0.07 -0.54 -0.40 -0.06 -0.01 -0.50
α SE (0.48) (0.35) (0.43) (0.53) (0.43) (0.35) (0.34) (0.45) (0.39) (0.28) (0.31) (0.41)
MF β 1.25*** 1.06*** 1.04*** 1.17*** 1.28*** 1.11*** 1.07*** 1.22*** 1.33*** 1.12*** 1.11*** 1.27***
β SE (0.09) (0.06) (0.08) (0.10) (0.08) (0.05) (0.06) (0.09) (0.09) (0.05) (0.05) (0.08)
6 SM B5 0.29* 0.97*** 1.01*** 0.49*** 0.22 0.91*** 0.96*** 0.35** 0.12 0.87*** 0.91*** 0.24*
SM B5 SE (0.16) (0.10) (0.12) (0.18) (0.15) (0.09) (0.10) (0.15) (0.15) (0.08) (0.09) (0.14)
W ML 0.64*** 0.47*** 0.44*** 0.58*** 0.69*** 0.59*** 0.51*** 0.65*** 0.77*** 0.57*** 0.56*** 0.71***
W M L SE (0.17) (0.11) (0.13) (0.20) (0.17) (0.12) (0.11) (0.19) (0.16) (0.10) (0.11) (0.17)
Adj. R-squared 0.63 0.73 0.63 0.53 0.71 0.69 0.74 0.65 0.76 0.79 0.75 0.72
Significance levels are marked as *** p<0.01, ** p<0.05, * p<0.1

Source: Calculations by the author using LSEG Datastream data and monthly factors and risk-free rate from
Raju (2022b), available at https://www.invespar.com/research
Notes: The table reports the intercept and intercept, MF, SMB, and WML coefficients of the 6-factor regression
model (Equation 3).
January 2024 24

SSRN Id4687044

Uploaded by

Copyright:

Available Formats

SSRN Id4687044

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

SSRN Id4687044

Uploaded by

Copyright:

Available Formats

Timing the Tide: The Impact of Rebalancing Periods in

Momentum Investing in Indian Equities

Keywords— Factors, Portfolio Construction, Momentum, Rebalancing, Indian Equity

JEL Classification Codes G00, G11, C15

Electronic copy available at: https://ssrn.com/abstract=4687044

effect of rebalancing frequency on momentum portfolios in Indian equities.

understanding momentum investing in India’s distinct market.

Section 3 presents our results; and we conclude in Section 4.

2.1 Portfolio Construction

(equal weight and rank weight).

• Market Cap Based Weighting Schemes

weighted proportionally on this normalised z-score, Z, wi = Z

• Non-market-cap-based Weighting Schemes

– Equal Weighting: Each firm has the same weight, wi = N.

and then weighted proportionally, wi = PN

Through this method, we create 4 (rebalancing frequency) x 3 (universe) x 3 (portfolio holdings) x 4

where the rebalance frequency is 6 months.

craftsmanship alpha in the Indian equity market context.

2.2 Summary Statistics

formula, we test the significance of differences in Sharpe ratios:

Effective holdings are calculated using the Herfindahl index:

and an average for the period is taken.

2.3 Returns-based Analysis

the Indian market:

Ri,t = α + βM F + βSM B SM B5 + βHM L HM L + βRM W RM W + βCM A CM A + βW M L W M L + εi,t (3)

the residual error term for period t.

3 Results and Discussion

performance characteristics of various levers in implementing a momentum strategy in India.

3.1 Summary Statistics

portfolio performance under different scenarios.

schemes, underscoring the temporal nature of momentum.

size for momentum strategies.

fication and concentrated exposure to capture momentum.

balancing frequencies we test.

3.2 Sharpe and Sortino Ratios Analysis

how the risk-return trade-off varies across different portfolio strategies.

momentum effect more robustly.

the importance of periodic portfolio adjustments in momentum investing.

with larger holdings.

3.3 Drawdown Measures Analysis

particularly in the context of momentum investing.

1. Impact of Rebalancing Frequency on Maximum Drawdown: The data suggest a notable

ing the universe to the top 200.

a 1-month rebalancing period.

frequent rebalancing in risk management.

the top 500 and 750 stocks.

reflecting the amplified risk associated with a broader market exposure.

drawdowns, indicating a higher volatility and risk profile.

3.4 Additional Portfolio Characteristics Analysis

allows winners to compound their returns.

2. Effective Holdings and Diversification Trends: Effective holdings, indicative of diversification,

8.30 in a 6-month rebalancing, suggesting a slight reduction in diversification as the frequency of

rebalancing decreases, potentially increasing the risk of the portfolio.

reaction to market changes compared to smaller stocks.

4. Portfolio Turnover across Rebalancing Frequencies: Portfolio turnover is a crucial metric in

centage of new entrants.

the importance of the rebalancing frequency.

percentage of total turnover.

contribution of the increase in the weights of the continuing stocks decreases.

Significance levels are marked as * p<0.01, p<0.05, * p<0.1