by Mayank Gupta, Satyam Kumar, Abhinandan Borad, Subrat Kumar Seet and Pratibha Kedia^{^}

This article examines the determinants of equity mutual fund flows in India, highlighting the shift in retail investor behaviour toward equity-oriented assets over the past decade. Using machine learning techniques, the empirical analysis identifies increasing financial inclusion (proxied by demat accounts), fixed deposit rates, and business confidence as the top three influential factors shaping equity mutual fund flows. Granger causality analysis further suggests that real GDP growth helps forecast these flows, reflecting how economic performance influences investor behaviour. The article underscores the growing maturity and long-term orientation of retail investors in India. Despite recent growth, India’s mutual fund industry remains relatively small compared to those of advanced economies, indicating significant scope for future expansion.

Introduction

Retail investors have been increasingly preferring equity investments over traditional saving instruments, which is reshaping the financial landscape in India. The substantial growth of mutual funds (MFs) would unquestionably be a prominent highlight if one were to outline the significant developments in India’s financial sector during the twenty-first century. The share of MFs in the household sector’s gross financial savings increased from 0.9 per cent in 2011-12 to 6 per cent in 2022- 23 (Chart 1a). MFs have emerged as the preferred vehicle for household investors to invest in equity markets¹ (Chart 1b).

Over the past few decades, MFs in India have experienced considerable popularity, attributable to rise in income levels, increasing levels of financial literacy, young demographic composition, the widespread growth of the digital ecosystem and internet connectivity, and the success of the marketing initiatives led by the Association of Mutual Funds in India (AMFI) leading into buildup of trust. Assets under management (AUM) of the MF industry have grown from ₹6.1 lakh crore at end-March 2010 to ₹65.7 lakh crore at end-March 2025 at a compounded annual growth rate (CAGR) of 17.1 per cent. Monthly flow through systematic investment plans (SIPs) has been recording fresh lifetime highs, crossing the ₹27,000 crore mark in June 2025, despite heightened volatility in Indian equity markets in the recent period due to geopolitical developments and trade related uncertainties.

MFs are seeing higher growth in new SIP accounts from smaller towns (beyond 30 or B-30 centers) compared to the top 30 (T30²) cities³ along with greater participation from women - their share in industry assets has expanded from 15 per cent in March 2017 to nearly 21 per cent as of December 2023. Another encouraging facet of this change has been the fact that the pace of growth of women’s participation is more prominent in smaller towns - the share of women’s folios and assets in B-30 cities has increased from 15 per cent and 17 per cent to 18 per cent and 28 per cent, respectively, during the same period⁴. The AUM of women investors in MFs in India more than doubled from ₹4.59 lakh crore in March 2019 to ₹11.25 lakh crore in March 2024⁵. At a time when foreign portfolio investor’s (FPIs) share in the equity market is decelerating albeit slowly, the ownership of MFs in domestic equity markets is at an all-time high. This trend is expected to strengthen⁶, given the rapid projected growth of the domestic MF industry in the coming years. In such an environment, it is imperative to closely examine the factors influencing the inflow of funds into MFs, particularly into equity-oriented MFs, which have attracted substantial interest from retail investors and the subsequent implications for the domestic financial markets.

Against this background, the present study attempts to investigate the key determinants of flows into equity MFs in India using machine learning techniques like random forest. When the data exhibits non-linear and interacting effects, random forest is capable of managing complex interactions between features much better than linear models. This exercise involves exploring various potential channels, such as the comparative returns on alternative investment avenues, increasing financialisation of savings, and the overall outlook on the financial markets and business environment.

The rest of the article is organised as follows. Section II discusses stylised facts on equity MFs and recent trends. Section III reviews the extant literature on equity MFs. Section IV covers the analysis of determinants of flows into equity MFs and an assessment of the link between MF flows and real Gross Domestic Product (GDP). Lastly, section V presents the concluding observations.

II. Stylised Facts of the MF Industry in India

The increasing interest in MFs can be gauged by the rise in the number of investor accounts, which have grown nearly six times since December 2014, reaching 23.5 crores in March 2025. Notably, 91. 7 per cent of these accounts belong to retail investors (Charts 2a and 2b). The number of unique MF investors crossed the five-crore mark in 2024⁷.

MFs are emerging as a potential competitor to bank deposits in terms of investment avenue, especially among the aspirational middle class in India. The ratio of AUM of the MF industry to total deposits⁸ has more than doubled from around 10 per cent at end-March 2014 to 23.8 per cent at end-March 2024 (Chart 3a). Despite the recent high-growth phase, MFs still have a vast territory to capture. MFs’ assets in advanced economies have a higher share as a per cent of their GDPs than in India (Chart 3b).

Equity MFs have grown faster than non-equity funds, supported by the stellar performance of domestic equity markets and sustained inflows into equity-oriented funds from retail investors, with equity AUM rising from ₹2.1 lakh crore at end-March 2010 to ₹34.5 lakh crore at end-March 2025 (Charts 4a and 4b).

The tilt of retail investors towards equities is also reflected in the fact that around two-thirds of all MF accounts are focused on equity-oriented assets. There has been a simultaneous growth in SIPs along with the rise in the number of investor accounts. Indian MFs had almost 9.2 crore SIP accounts at the end of June 2025 through which investors regularly invested in Indian MF schemes. While growth in the number of investor accounts indicates the increasing popularity of MFs among small investors, a secular increase in SIP investments is a testament to the growing maturity of retail investors (Chart 5a). Two polarising trends i.e., preference for short-term trading via equity derivatives and long-term investments through SIPs have gained traction simultaneously. It is noteworthy that while there have been concerns over retail investors being on the losing side of derivatives trading⁹, a segment of retail investors has shown more patience, which is reflected by the fact that retail investors hold 61 per cent of equity MF assets for a period greater than 24 months (as at end-March 2025) (Chart 5b). The stickiness of retail investors with equities is showing signs of strengthening as this share (retail investors holding equity assets for more than two years) stood at 44.9 per cent and 53.3 per cent in end-March 2023 and end-March 2024, respectively. MFs seem to have successfully persuaded small investors to stay invested for the longer term.

As discussed above, there are signs of increasing preference for equities beyond metro cities, as seen by the rise in the number of fresh demat account openings from tier-2 and tier-3 towns¹⁰. Although small cities are in the early stages of developing a taste for equities, and much of the volume still originates from a handful of states, visible changes can be seen. Regarding equity-oriented MFs, Maharashtra alone accounts for over a quarter of the AUM. Densely populated northern and eastern states account for a relatively lower share (Chart 6a). However, a glance at the growth rate in AUM shows that laggard states have shown a tendency to catch up (Chart 6b).

The market regulator and the industry body have proactively promoted the deepening of MFs beyond T30 cities. As of end-June 2025, 18 per cent of the MF industry’s assets came from B30 locations. MFs could charge an additional expense ratio of 0.3 per cent on fresh inflows of up to ₹2 lakhs from B30 cities to provide distributors a higher commission to popularise MFs in smaller towns¹¹. Despite the growth of digital platforms in recent years that enable direct investments in MFs, retail investors tend to invest through distributors who provide administrative assistance and guidance. Only 27 per cent of the retail investors chose to invest directly (29 per cent of equity scheme assets came through the direct route as of end-June 2025). Corporate houses and HNIs are concentrated in large cities, and thus, T30 locations account for most of the non-equity assets (Chart 7).

Equities are the most significant investment class for MFs, with the banking and finance sector accounting for more than one-fourth of MF equity investments (Chart 8).

With increasing flows towards MFs, their ownership in domestic equities has grown considerably over the last decade. The shareholding of MFs in companies listed on the National Stock Exchange (NSE) has risen from 3.7 per cent at end-March 2010 to 10.4 per cent at end-March 2025 (Chart 9).

With increasing shareholding in companies and having substantial voting power, MFs have a greater responsibility to ensure that the companies they invest in, abide by corporate governance principles. To ensure that MFs do not shy away from their stewardship role and vote in the best interest of their unitholders, effective April 1, 2021, SEBI has mandated MFs to vote on several resolutions ranging from corporate governance to changes in capital structure¹² of the companies they invest in. Following the change in guidelines, MFs have shown more active and diverse participation in voting on corporate resolutions, which is also reflected in a greater share of voting ‘Against’ the resolution. Share of MFs voting ‘Against’ a corporate resolution has increased from 4.4 per cent in 2014-15 to 11.9 per cent in 2024-25 while MFs abstaining from voting on resolutions has declined from 21.5 per cent in 2014-15 to 0.2 per cent in 2024-25.

Another notable trend has been the growing interest in passive investments worldwide. As per Morningstar, passive investments represent 43.5 per cent of worldwide long-term assets as of 2024, a 3.2 per cent increase from year-end 2023¹³. Passive funds have also found favour amongst domestic investors, with the share of passive funds as a per cent of industry assets rising from 6.1 per cent in March 2019 to 17 per cent in March 2024¹⁴. During 2024-25, passive funds/Exchange Traded Funds (ETFs) witnessed net inflows of ₹1.4 lakh crore, led by inflows in domestic ETFs and domestic index funds. Equity-oriented ETFs have the largest share supported by institutional flows¹⁵ (Chart 10).

MFs also provide opportunities for domestic investors to get exposure to international equities and reap diversification benefits. The trend of looking beyond domestic equities picked up during the COVID-19 pandemic. However, it has moderated in recent periods amid outperformance by domestic equity markets coupled with the MF industry nearing the maximum permissible ceiling limit for overseas investments (Chart 11a). Restrictions on overseas investing via MFs may inadvertently push investors towards alternate routes, such as the Liberalised Remittance Scheme (LRS), leading to potentially riskier exposures in international equities (Chart 11b).

The re-categorisation of MF schemes has provided greater clarity to investors about the schemes’ final investment universe. Flexi-cap¹⁶ funds have witnessed substantial net inflows among open-ended equity schemes, likely due to the greater flexibility provided by these schemes, while sectoral/thematic funds¹⁷ gained traction in 2024-25 (Chart 12).

Sustained MF inflows have helped cushion the equity market against volatility triggered by FPI outflows. The capacity of MFs to influence domestic markets has strengthened. Empirical evidence suggests that MF flows affect equity market returns in the small and midcap segments¹⁸. The broader markets have witnessed increasing investor interest over the last few years, with midcap and smallcap funds accounting for almost one-fifth of the net inflows into open-ended equity-oriented MFs in 2024-25 (Chart 13). The market value of Small and Midcap stocks held by MFs stood at around ₹14 lakh crore¹⁹ in end-March 2025, accounting for more than one-fourth of the equity holdings of MFs. Given relatively low liquidity in the Small and Midcap segments, MFs could be subject to large liquidity risks from redemption pressures in case of sharp downward adjustments, which could then put pressure on other segments of the financial markets. To protect investor interest of small and midcap funds, SEBI has asked MFs to put in place appropriate and proactive measures to protect investors by using tools like moderating inflows, portfolio rebalancing and ensuring that investors are protected from the first mover advantage of redeeming investors in case of mass redemptions. Furthermore, the market regulator has regularised liquidity stress tests for these equity schemes, an exercise already conducted regularly for debt schemes.²⁰

III. Literature Review

Understanding the factors that influence flows into mutual funds is important because these flows reflect consumer saving patterns, individual wealth, and investment decisions. They also impact fund managers’ incentives, reveal investor behaviour, and influence the overall efficiency of financial markets (Ferson and Kim, 2012; Kopsch et al., (2015)). Past studies on MF flows have employed both micro and macro approaches. The micro approach focuses on analysing flows into individual MFs, while the macro approach examines the aggregate inflows and outflows across the MF industry as a whole (Alexakis et al., 2005,2013; Remolona et al., 1997; Edward and Zhang, 1998; Watson and Wickramanayke, 2012; Jank, 2012; Fong et al., 2018).

In the Indian context, studies have focused on analysing flows into individual funds as well as overall flows in the MF industry in the case of both debt and equity funds (see Kumar et al., 2020; Mishra, 2011; Gupta et al., 2022). Madhumathi et al., (2012) examines the factors influencing equity MF flows in India. This study covers the period from July 2005 to August 2012, uses linear AutoRegressive Integrated Moving Average (ARIMA) models, and finds volatility, volume, dividend yield, exchange rate, investor behaviour, US market returns, call rates and past fund flows as factors influencing net flows in equity MFs. A recent study by Kumari and Debnath (2022) investigates the determinants of MF flows with reference to BSE SENSEX using monthly data from January 2012 to December 2020. The authors find that stock returns and volatility have an asymmetric impact on MF flows and foreign institutional investors (FIIs) have opposite trading pattern i.e., FII purchases lead to MF sales and vice versa.

In recent years, equity MF schemes in India have consistently seen higher net inflows than debt schemes since 2017, except in 2020, indicating a shift in investor preference. For the current study, the macro approach is chosen to gauge general shifts in investor preference and behaviour rather than specific fund-to-fund shifts by looking into the determinants of equity MF flows.

IV. Data and Methodology

The following variables with a monthly frequency are used to estimate the drivers of flows into equity MFs (Table 1). The data spans from April 2015 to November 2024.

Given the diverse and dynamic nature of factors influencing mutual fund flows, machine learning algorithms are well-suited for uncovering the most influential predictors. Unlike traditional linear models, which may miss complex interactions or non-linear patterns in the data, machine learning algorithm such as random forest can effectively handle such complexities without assuming a specific functional form. Random Forest is an ensemble machine-learning method constructed based on individual tree classification or regression (Breiman 2001). Ensemble learning combines multiple models to create a more robust and accurate predictive model. In the case of random forest, individual models are decision trees. It is a non-parametric supervised machine learning algorithm and reduces variance, resulting in a stable and reliable model. It is robust to outliers in the data due to the nature of averaging predictions from multiple trees and can also handle missing values.

The importance of variables is a crucial aspect of the random forest model, offering insights into the contribution of each feature to predictive performance. To evaluate variable importance, we employ both the permutation test and Shapley values. The permutation test involves randomly shuffling the values of a specific predictor to break its relationship with the outcome variable, while keeping all other variables unchanged. The resulting increase in prediction error indicates the importance of that variable-larger increases suggest a greater contribution to the model’s accuracy. In contrast, Shapley values, derived from cooperative game theory, measure each feature’s average marginal contribution to the prediction across all possible combinations of features. This approach provides a consistent and locally accurate estimate of variable importance, effectively capturing interactions and dependencies among features.

To understand the determinants of equity MF flows, variables are chosen based on cues from Kopsch et al., (2015) as well as other domestic factors such as fixed deposit rates, business confidence index and demat account openings. Fixed deposit rates are chosen as a substitute for the return/relative market return from equity MFs. A persistently low fixed deposit rate for an extended period might eventually lead people to search for other asset classes that offer higher returns, thereby increasing equity MF flows. The return on BSE Sensex has beaten fixed deposit returns in 10 out of the last 15 financial years with the Compounded Annual Growth Rate (CAGR) of Sensex Total Return Index at 12 per cent vis-à-vis 7 per cent for term deposits between 2010-11 and 2024-25 (Chart 14).

Demat account opening can be considered a proxy for financial inclusion and increasing preference for equities. This should lead to additional flows to equity-oriented products. The business confidence index is expected to impact flows, as it is an indicator of future growth, and the VIX is associated with volatility. Thus, higher volatility may reduce flows into equity MFs. According to Jank (2012), higher term spreads and risk spreads tend to result in lower flows in equity MFs. The author notes that just before or during a recession, there is a surge in dividend yield due to falling stock prices. Simultaneously, investors become less willing to hold equity positions. Therefore, a negative relation between dividend yield and flows in equity MFs might be observed. The direction of flows with inflation is not certain. Equities provide a hedge against inflation in emerging market economies (Al-Nassar and Bhatti, 2019; Spyrou, 2019). However, higher inflation might lead people to hedge by investing in real assets rather than stocks (Park et al., 1990; Kopsch et al., 2015). In this analysis, using both the permutation test and Shapley values, the primary focus is on identifying which variables significantly influence mutual fund flows, rather than determining the direction (positive or negative) of their effects. The objective is to assess the relative importance of each predictor in driving model performance. The analysis²¹ reveal the top five variables ranked by importance (Chart 15).

The results from both the permutation test and Shapley values indicate that demat accounts, fixed deposit rates, and the business confidence index are the top three influential predictors of equity mutual fund flows. This suggests that investor access to markets (as proxied by demat accounts), alternative saving options (such as fixed deposit rates), and optimism about the business environment (as measured by the business confidence index) play a significant role in driving flows to equity mutual funds. While both methods broadly agree on the most important variables, the permutation test highlights dividend yields as an important factor, whereas Shapley values emphasise term spreads. This divergence reflects how different methods capture variable importance: permutation importance focuses on the marginal impact on prediction accuracy, while Shapley values also account for interactions and shared contributions among variables.

Further, to understand the link between flows into equity MFs and the dynamics of the real economy i.e., whether equity MF flows contain information about real GDP growth or vice versa in the case of India, an empirical exercise is undertaken, and the analysis suggests that equity MF flows do not contain any information about real GDP growth²². The period of analysis is from Q1: 2012-13 to Q3: 2024-25 (Chart 16)²³.

A reduced-form bivariate VAR model is used. The results of the Granger Causality test between flows and real GDP growth are reported in Table 2. The null hypothesis that “flows do not Granger-cause real GDP growth” fails to be rejected, suggesting that the past value of flows does not contain significant information about real GDP growth. However, the null hypothesis “Real GDP growth does not Granger-cause flows” is rejected at five per cent level of significance. Thus, past values of real GDP growth contain useful information for forecasting equity mutual fund flows. A unidirectional Granger causality is established from real GDP growth to mutual fund flows. In essence, stronger real economic growth enhances investor’s financial capacity and confidence, enabling greater participation in equity markets.

V. Conclusion

There is an increasing preference for equity investments in India, particularly through equity MFs, which are experiencing continuous growth and resilience. Despite a traditionally cautious approach to financial markets, the populace has shown a discernible inclination to embrace greater financial risk in their personal investment choices in recent years. MFs have notably emerged as a favoured instrument in facilitating and accommodating this shift in financial behaviour.

The empirical analysis highlights that increasing financial inclusion (proxied by demat accounts), fixed deposit rates, and business confidence are the top three influential factors shaping equity mutual fund flows in India. An alternative analysis suggests that equity MF flows do not predict real GDP growth, while real GDP growth does help forecast flows. This indicates that stronger economic growth enhances both the capacity and confidence of investors to allocate more funds toward equity markets.

In an evolving scenario, keeping a close eye on the drivers of equity MFs may be necessary, as they have implications for household savings and the changing dynamics of domestic capital markets. Increasing retail participation, both directly and indirectly, also warrants more efforts toward investor education and protection to maintain the faith and trust of these new entrants. As MFs grow in size, a constant monitoring of risks emanating from their operations would need greater attention.

References

Alexakis, C., Niarchos, N., Patra, T. and Poshakwale, S. (2005), “The dynamics between stock returns and mutual fund flows empirical evidence from the Greek market,” International Review of Financial Analysis, Vol. 14, No. 5, pp. 559-569.

Alexakis, C., Apostolos, D. and Grose, C. (2013), “Asymmetric dynamic relations between stock prices and mutual fund units in Japan. An application of hidden cointegration technique”, International Review of Financial Analysis, Vol. 28, pp. 1-8.

Al-Nassar, N. S., and Bhatti, R. H. (2019). Are common stocks a hedge against inflation in emerging markets? Journal of Economics and Finance, 43, 421-455.

Breiman, L. (2001). Random forests. Machine learning, 45(1), 5-32.

Edwards, F. R., and Zhang, X. (1998). Mutual funds and stock and bond market stability. Journal of Financial Services Research, 13(3), 257-282.

Ferson, W.E. and Kim, M.S. (2012). “The factor structure of mutual fund flows,” International Journal of Portfolio Analysis and Management, Vol. 1, No. 2, pp. 112-143.

Fong, T. P. W., Sze, A. K. W., and Ho, E. H. C. (2018). Determinants of equity mutual fund flows–Evidence from the fund flow dynamics between Hong Kong and global markets. Journal of International Financial Markets, Institutions and Money, 57, 231-247.

Gupta, M., Kumar, S., Seet, S., and Borad, A. (2022). Market Returns and Flows to Debt Mutual Funds. RBI Bulletin October 2022, Reserve Bank of India.

Jank, S. (2012). Mutual fund flows, expected returns, and the real economy. Journal of Banking and Finance, 36(11), 3060-3070.

Kopsch, F., Song, H. S., and Wilhelmsson, M. (2015). Determinants of mutual fund flows. Managerial Finance, 41(1), 10-25.

Kumar, P., Saxena, C., and Gupta, A. K. (2020). A study on relationship between stock market returns and mutual fund flows. Journal of Commerce and Accounting Research, 9(1), 1.

Kumari, M., and Debnath, P. (2022). Determinants of Mutual Fund Flows in the Indian Stock Market: Insights Through the Nonlinear Autoregressive Distributed Lag Bounds-testing Approach with Structural Breaks. Vision.

Madhumathi, R., Gopal, N., and Ranganatham, M. (2012). Determinants of Debt and Equity Mutual Fund Flows in India. In XI Capital Markets Conference (pp. 21-22).

Mishra, P. K. (2011). Dynamics of the relationship between Mutual Funds investment flow and stock market returns in India. Vision, 15(1), 31-40.

Park, J.Y., Mullineaux, D.J. and Chew, I.-T. (1990), “Are REITs inflation hedges?”, Journal of Real Estate Finance and Economics, Vol. 3, No. 1, pp. 91-103.

Remolona, E. M., Kleiman, P., and Gruenstein Bocain, D. (1997). Market returns and mutual fund flows. Economic Policy Review, 3(2), 33-52.

Securities and Exchange Board of India. (2024). Working paper on household savings through Indian securities market. Department of Economic and Policy Analysis. https://www.sebi.gov.in/reportsand-statistics/research/sep-2024/working-paper-on-household-savingsthrough-indian-securities-market_86459.html

Spyrou, S. I. (2004). Are stocks a good hedge against inflation? Evidence from emerging markets. Applied Economics, 36(1), 41-48.

Watson, J. and Wickramanayake, J. (2012), “The relationship between aggregate managed fund flows and share market returns in Australia,” Journal of International Financial Markets, Institutions and Money, Vol. 22, No. 3, pp. 451-472.

^{^} The authors are from the Department of Economic and Policy Research. The views expressed in this article are those of the authors and do not represent the views of the Reserve Bank of India.

¹ The current Household Financial Savings Data accounts only for primary market flows into equities. Securities and Exchange Board of India (2024) undertook an extended analysis by incorporating secondary market flows and introducing certain methodological refinements. Even with these adjustments, the findings indicate that flows into mutual funds continue to significantly surpass the combined flows into equities through both primary and secondary markets.

² T30 refers to the top 30 geographical locations in India and B30 refers to the locations beyond the top 30. The list of T30 cities can be accessed at https://www.amfiindia.com/research-information/aum-data/listoftop30cities

³ https://www.business-standard.com/markets/stock-marketnews/sip-soars-beyond-city-limits-small-towns-lead-in-account-additions-123112600539_1.html

⁴ https://www.amfiindia.com/Themes/Theme1/downloads/AMFI_womensDay_Mar2024.pdf

⁵ https://www.amfiindia.com/Themes/Theme1/downloads/AMFIFactbook%202024.pdf

⁶ https://www.mordorintelligence.com/industry-reports/india-mutual-fund-industry

⁷ https://www.business-standard.com/markets/mutual-fund/mf-investor-count-crosses-50-million-after-10-million-additions-in-a-year-124102501213_1.html

⁸ Deposits include deposits with scheduled commercial banks (SCBs), public deposits with non-banking financial companies (NBFCs) and urban-co-operative banks.

⁹ https://www.sebi.gov.in/reports-and-statistics/research/jul-2025/comparative-study-of-growth-in-equity-derivatives-segment-vis-vis-cash-market-after-recent-measures_95105.html

¹⁰ The data may slightly overstate the trend due to migrant workers not updating their addresses.

¹¹ SEBI has stopped the incentive scheme from March 1, 2023, till further directions due to misuse, inconsistencies and deficiencies in the manner of implementation of the incentive mechanism.

¹² https://www.sebi.gov.in/legal/circulars/mar-2021/circular-on-guidelines-for-votes-cast-by-mutual-funds_49405.html

¹³ https://www.morningstar.com/business/insights/research/global-asset-flows-report

¹⁴ https://www.amfiindia.com/Themes/Theme1/downloads/AMFIFactbook%202024.pdf

¹⁵ EPFO invests in equity markets through ETFs replicating BSE-SENSEX and NIFTY-50 indices.

¹⁶ SEBI introduced this category in November 2020 as an an open-ended dynamic equity scheme investing across large cap, mid cap, small cap stocks.

¹⁷ https://www.cnbctv18.com/personal-finance/mutual-funds-thematic-schemes-rise-in-inflows-returns-should-you-invest-19424553.htm

¹⁸ https://rbi.org.in/web/rbi/-/publications/rbi-bulletin/state-of-the-economy-22189?category=23918848

¹⁹ As per calculations using data from Prime MF database.

²⁰ https://www.msn.com/en-in/news/other/latest-stress-test-shows-this-about-small-cap-mutual-funds-check-details-here/ar-AA1FMddc?ocid=BingNewsSerp

²¹ The dataset is standardised to have a mean of zero and a standard deviation of one. An 80–20 split is then employed to divide the data into training and testing sets.

²² Net flows are normalised by the previous month’s net AUM to control the increasing trend in flows (Remolona et al., 1997).

²³ The correlation coefficient between the two series is 0.17.