Does investor sentiment affect the Indian stock market? Evidence from Nifty 500 and other selected sectoral indices

Abstract

Investor sentiment is the result of irrational speculations about the future asset values driven by the market participants. Though scholarly works on investor sentiment are evolving in both developed and emerging markets, the literature in the Indian context is relatively modest. To fill this void, the study aims to examine sentiment-return relations based on a sectoral analysis framework. The study considered Nifty 500 and sectoral indices return such as Automobile, Information Technology, Metal, Fast-Moving Consumer Goods, and Public Sector Undertakings. To measure investor sentiment a unique sentiment index (INDex) using seven indirect proxy sentiment indicators suitable for the Indian stock market is proposed. To test the framed hypothesis, the study employs Principal Component Analysis and OLS regression. The results uncover that there exists a strong significant positive sentiment effect on Nifty 500 and selected sectoral indices return. The findings assist academicians, practitioners, investors, and policymakers in enhancing their understanding of the sentiment-return nexus in the Indian stock market and thereby guide them to ensure caution while making investment decisions.

Keywords:

• Investor sentiment
• stock return
• principal component analysis
• OLS regression model
• India

REVIEWING EDITOR:

• Xibin Zhang, Monash University, Australia

SUBJECTS:

• Economics
• Finance
• Business, Management and Accounting

1. Introduction

In the last decade, scholarly interest in investor sentiment (IS) and its impact on stock return (SR) is flourishing (Baker & Wurgler, 2007; De Long et al., 1990; Wang et al., 2022; Zhang et al., 2018). Sentiment refers to investors’ emotions and irrationality that influence their decision-making and explains the market movements, particularly at the time of market crashes and pandemics (Parveen et al., 2023). Brown and Cliff (2004) stated that ‘investor sentiment is formed by the overly optimistic (high sentiment) or pessimistic (low sentiment) attitudes of speculative investors’. Baker and Wurgler (2007) defined IS as a ‘belief about future cash flows and investment risks that are not justified by the facts’. Therefore, sentiment plays a crucial role in explaining asset prices, SR, and volatility (Baker & Wurgler, 2007; De Long et al., 1990). As sentiments cannot be quantified directly (Aggarwal & Mohanty, 2018), researchers have constructed sentiment indices using various measures to examine their effect on return and volatility (Andleeb & Hassan, 2023; Baker & Wurgler, 2006). These measures include survey measures, market measures, and media measures.

As far as scholarly works are concerned with IS and SR nexus, one facet of the literature shows that SR is significantly affected by the trading behavior of market participants (Baker & Wurgler, 2006; Dergiades, 2012). On the contrary, the literature provides evidence for a negative sentiment-return relationship (Brown & Cliff, 2005; Fisher & Statman, 2000). Further studies also observed an insignificant relationship between IS and SR (Bathia et al., 2016; Kandır & Yücel, 2019). Notably, most of the studies on the sentiment-return relationship have been conducted in developed countries (Beer & Zouaoui, 2012; Bouteska & Mili, 2023; Kim et al., 2014; Li et al., 2017; Sakariyahu et al., 2023). However, the generalization of the facts from developed markets to emerging markets is challenging due to several factors such as uncertainty, ambiguity, and lack of sophisticated technology. With this view, a few studies have also investigated the impact of IS on SR in emerging markets (Andleeb & Hassan, 2023; Chen et al., 2013; Corredor et al., 2015; Vuong & Suzuki, 2020).

Though scholarly works on sentiment-return relations are increasing in emerging markets, the research has gained less attention from scholars and academicians in India (Kamath et al., 2022; Prasad et al., 2022). Further, it is demonstrated that the direct impact of individual and institutional investor sentiment on SR led to volatile market conditions in India (Priyesh & Jijo, 2023). However, proper indicators to measure this sentiment are lacking (Prasad et al., 2022). Thus, to fill this void in the literature, the study aims to construct a unique sentiment index (INDex) using seven indirect sentiment proxies suitable for the Indian equity market.

The novelty of the study lies in its early attempt to examine the IS effect on SR. Further, the study applies a sectoral analysis framework by considering five sectoral indices of the National Stock Exchange (NSE) and provide additional evidence on the sentiment-return nexus. The rationale behind the choice of the Nifty 500 is that it represents 96.1% of free float market capitalization. The results reveal that there exists a significant positive effect of IS on Nifty 500 and sectoral index returns. In the selected sectoral indices, Metal has the highest and Fast-Moving Consumer Goods (FMCG) has the lowest influence of IS compared to the other sectors. Therefore, the study adds contributions to the existing body of knowledge and thereby enhances the understanding of the impact of IS on the Indian equity market and guides various market participants, particularly investors.

The remainder of the paper is organized as follows. Section 2 provides the relevant literature and hypothesis for the study. Section 3 explains the methodology, variables considered, and construction of the sentiment index. Section 4 details the empirical results. Section 5 provides the discussion, and finally, Section 6 concludes the findings and provides scope for future research.

2. Literature review and hypothesis development

IS refers to the overall expectation and subjective view of investors towards the stock market. As sentiments cannot be measured directly, researchers have used different methods to measure their impact on SR, with both exploratory (Aggarwal & Mohanty, 2018; Ajlouni & Alghusin, 2021) and empirical analysis (Chen et al., 2020; Ding et al., 2019). The measures of IS are mainly divided into three categories such as survey, market, and media measures as seen in the literature. Survey measures include Investors’ Intelligence (Lee et al., 2002), the Wall Street Analyst Sentiment Index (Fisher & Statman, 2000), and daily survey data (Kling & Gao, 2008). Market measures such as mutual fund flow (Chi et al., 2012), net new added accounts (Chu et al., 2016; Li & Zhang, 2008), and trading volume (Rizkiana et al., 2019; Uygur & Taş, 2014). Media measures include Google search volume (Rizkiana et al., 2019), social media posts (McGurk et al., 2020), and Twitter data (Pyo & Kim, 2021).

Baker and Wurgler (2006, 2007) stated that sentiment has predictive power for small, young, and volatile stocks. Stambaugh et al. (2012) found that when sentiments are high, overpricing may occur, and investors can earn profitability with long-short strategies. Huang (2015) by proposing a new IS index suggested that IS is significant not only for the cross-section of SR but also at the aggregate market level. IS is positively associated with returns and when investors are optimistic (pessimistic) they are likely to drive the price up (down) (Smales, 2017). It is also observed that IS has an explanatory power, where high sentiment leads to an increase in SR (Ryu et al., 2017). Literature also reckons that IS might lead to risk-taking behavior in investors, where fear and uncertainty in investors provoke volatility in the market (Huynh et al., 2021). Further, it is found that the greater the level of panic, the greater the fall in SR, which clearly explains the significance of sentiment-return relations in the market (Aggarwal et al., 2021). A few studies in the developed market context have found negative (Brown & Cliff, 2004; Wang et al., 2022) and insignificant effect (Bathia et al., 2016; Kandır & Yücel, 2019) of IS on SR. Further, Kadilli (2015) states that IS is negative (insignificant) during normal periods and positive (significant) during crisis periods in developed countries. Interestingly, Dash and Mahakud (2012) constructed an aggregate sentiment index for the Indian market and found that stocks that are hard to value and arbitrage have a significant sentiment influence. Dash and Maitra (2018) found a strong effect while studying the relationship between IS and SR in the Indian stock market. A similar study found a significant positive correlation between sentiment index and SR (Aggarwal & Mohanty, 2018). On the contrary, sentiment index constructed by Mathur and Rastogi (2018) using indirect market sentiment proxies found that broad market returns are not predictable by sentiment. It is observed that the literature on the influence of IS on SR has shown mixed results both in developed and emerging markets.

It is well known that previous studies on the impact of IS on SR are mainly examined in the developed market context (Aboura, 2016; Chen & Lien, 2017), which creates significant importance for the studies in the emerging market context as the markets are prone to irrational behavior of investors. Further, it is observed that only limited empirical pieces of evidence are available in the Indian context (Dash & Maitra, 2018; Eachempati & Srivastava, 2021; Kamath et al., 2022; Prasad et al., 2022). Thus, this inconclusiveness has led the researchers to further investigate the effect of IS on SR in the Indian context. Moreover, past scholarly works focused only on the Nifty 50 and BSE Sensex indices (Eachempati & Srivastava, 2021; Yadav & Chakraborty, 2022). However, the proper knowledge of the potential effect of sentiment on different sectors in the Indian equity market is lacking. Further, in the notion of the remarkable influence of sentiment, the study predicted that there is a significant influence of investor sentiment on various indices. Therefore, this study intends to examine the effect of IS on SR of Nifty 500 and other sectoral indices of NSE. The study based on theoretical background and literature framed the following hypothesis:

H₁: Investor sentiment has a significant positive impact on the stock return of the Nifty 500 and other NSE sectoral (Auto, FMCG, IT, Metal, and PSU) indices.

3. Methodology

3.1. Data

The study uses secondary data, where monthly returns and other firm-specific indicators of Nifty 500, and five sectoral indices such as Automobile (Auto), Information Technology (IT), Metal, FMCG, and Public Sector Undertakings (PSU) were considered based on their data availability to ensure completeness, uniformity, and consistency in the data. The data were retrieved from the NSE (www.nseindia.com) between January 2004 and December 2021. These sectors are categorized following the industry classification benchmark of the NSE, which is a widely accepted and recognized framework.

3.2. Econometric model

The study employed Principal Component Analysis (PCA) and ‘Newey and West’s’ (1987) Adjusted Ordinary Least Square (OLS) regression analysis to examine the impact of IS on SR of Nifty 500 and sectoral indices. Nifty and sectoral indices are outcome variables, INDex (sentiment index) is the independent variable. The study uses PCA following the widely accepted Baker and Wurgler’s (2006, 2007) approach for sentiment index construction. The study uses an Augmented Dickey-Fuller (ADF) unit root test and correlation matrix to check the stationarity and the relationship between the variables.

3.3. Variables used for construction of sentiment index

Based on the previous literature on IS and its effect on SR, the study considers a wide range of variables to perform PCA. Furthermore, as there is no theoretical justification for the number of variables to be considered for the construction of the sentiment index, a comprehensive set of seven indirect market-wide proxy sentiment indicators suitable for the Indian capital market is considered. To this end, the study employs a set of unique and frequently used indirect firm-specific and market-wide proxy sentiment measures such as Advance-Decline ratio (ADR), Money Flow Index (MFI), Number of Initial Public Offerings (NIPO), Price-to-Earnings ratio (PE), Price–to–Book ratio (PB), Relativity Strength Index (RSI), and Turnover (Turn). Further, various macro-economic variables and market variables such as Dividend Yield (DY), Exchange Rate (ER), Foreign Institutional Investors (FII), Treasury Bill (T-bill), and United States Economic Policy Uncertainty index (USEPU) were considered to remove redundant business cycle fluctuations. The data for all the considered variables are retrieved from the NSE (www.nseindia.com) and FRED (www.fred.stlouisfed.org) website.

3.3.1. Stock return and sectoral return

Return for Nifty 500, Auto, FMCG, IT, Metal, and PSU is calculated by taking the monthly closing price from January 2004 to December 2021. Monthly closing prices are converted into compounded log returns as, r_t = ln(P_t/P_t-1) Where r_t is the compounded return at time t and P_t and P_t-1 are the monthly stock index at the two successive months t and t-1, respectively.

3.3.2. Advance decline ratio

The ADR represents market sentiment and explains to investors if the stock is rising or falling. It is the ratio of the number of stocks whose closing price is greater than the previous day’s closing price to the number of stocks whose closing price is lower than the previous day’s closing price. ADR can confirm the price trends in the market indices, and when divergence occurs, it can also warn of reversals. Brown and Cliff (2004) stated that rising (declining) trends in ADR can be used to validate the market’s upward (downward) trend. ADR is expected to be positive since IS makes the market active. As a result, the ratio aids in the recognition of recent trends in the market which can be utilized as an indicator to measure IS (Baker & Wurgler, 2006; Feldman, 2010; Haritha & Uchil, 2020).

3.3.3. Money flow index

It is a volume-weighted measure that shows whether the market is overbought or oversold. A value of 80(20) indicates that the market is overbought (oversold). The MFI has a positive relation with IS. The study constructs an MFI using the following technique. First, the researcher defines ‘Typical Price’ as TP_t = C_t + L_t + H_t/3 Where, TP_t (Typical price) at time t is the average of C_t (closing price), L_t (lowest price), H_t (highest price). In the next step, money flow is computed which is the product of the typical price and its turnover. Money flow_t = TP_t* Turnover. In the next step, Money Ratio_t which is calculated as the ratio of the positive money flow_t to negative money flow_t. Money flow is positive if TP_t ≥ TP_t-1 and vice versa. At last, the study estimates the MFI as MFI_t = 100 – (100/1 - Money Ratio_t)

3.3.4. Number of initial public offerings

Baker et al. (2012) stated that a firm aims to get additional capital when a firm’s market value is high and repurchases its stocks when their market value is low. The purpose is to benefit from the sentiment in the market. According to the market timing hypothesis, a lower (higher) value of NIPOs means the market sentiment is bearish (bullish) (Baker & Wurgler, 2006). The NIPOs are calculated as the log of the total number of IPOs in a month i.e., ln(1 + NIPOs). Before log transformation, one is added to account for nil values. Baker and Wurgler (2006) and Brown and Cliff (2004) used this proxy variable as a sentiment indicator.

3.3.5. Price-to-book ratio

It is the ratio of share price to book value per share. A PB ratio is employed to identify undervalued stocks. A lower PB ratio indicates that the stock is undervalued, and a higher ratio indicates the stock is overvalued and tends to imply high confidence in the firm. IS is positively correlated with the PB ratio. Among market indicators, the PB ratio is frequently seen as a price-related indicator of IS (Brennan & Wang, 2010; Hirshleifer, 2001). Here PB reflects the firm’s book value in the market.

3.3.6. Price-to-earnings ratio

PE is the relative value of market price and earnings per share. It is regarded as one of the widely utilized indicators of stock valuation due to its intuitive appeal and practical simplicity. It usually reveals investors’ sentiment, if the ratio is high, investors will be optimistic and expect higher earnings growth in the future. If the PE ratio is lower, investors will be pessimistic about the market (Basu, 1977). Zhang et al. (2018) state that the PE ratio is a measure of investors’ forecasts of future stock values. Yang and Zhou (2015) have used the PE ratio as a proxy indicator of investor sentiment.

3.3.7. Relative strength index

RSI is a prominent market indicator for determining whether a market is overbought or oversold. An RSI of 80 indicates that the market is overbought and 20 indicates that the market is oversold. 14 days RSI is one of the popular RSI in the market. Yang and Zhou (2015) also employ this indicator in the construction of the composite sentiment index. In this study, RSI is calculated as RSI (14)_t = 100 ∑10_i (P_ti – P_ti – 1)+/∑10_i (P_ti – P_ti – 1), Where P_t is the closing price of the index I at time t and P_t-1 is the closing price of stock index I at time t-1.

3.3.8. Turnover

IS can be observed using Turn as it is considered a proxy of market liquidity. It measures the stock market trading activity (Chong et al., 2017). When irrational investors are optimistic, they participate in the market actively and expedite the turnover volume (Baker & Stein, 2004). Karpoff (1987) stated that Turn will be high in a bullish market and low in a bearish market. The Turn is proposed to increase when the sentiment of investors is strong, and it is proposed to decrease when the sentiment is weak, and it is a positive IS proxy (Baker & Wurgler, 2006). The low Turn is generally followed by a price decrease, while the high Turn is preceded by a price rise (Ying, 1966). Therefore, myriad researchers have used Turn as a proxy of IS (Baker et al., 2012; Chen et al., 2020). In the study, Turn is defined as the detrended natural log of the raw turnover.

Furthermore, the study employs the following macroeconomic variables as they stimulate economic growth and stability. Researchers have used FII, DY, T bill, USEPU, and ER (Aggarwal & Mohanty, 2018; Ghumro et al., 2022; Sibley et al., 2016). It presumes a positive relationship of the FII with the stock market as they demonstrate growth in the economy and a negative relationship of the DY, T bill, and ER (Aggarwal & Mohanty, 2018). The data for these have been taken from the NSE and RBI website (www.rbi.org.in). USEPU assesses market uncertainty, with globalization and trade linkages with the USA, increasing uncertainty in US markets unfavorably affects trade links and generates negative sentiments (Aggarwal & Mohanty, 2018).

3.4. Construction of sentiment index

It is observed that rational and irrational components of investors might influence the proxies of IS. To adjust and remove these business cycle fluctuations and their influence on sentiment, proxies are regressed to the various macroeconomic and market variables. The obtained residuals are used as orthogonal sentiment proxies for constructing a raw sentiment index by employing PCA.

PCA is a factor analysis that gives a small set of linear combinations from a large set of variables which accounts for the highest variance in the original set. The advantage of employing PCA is that it filters out the idiosyncratic noise in the orthogonalized sentiment measure and captures their common components. It has been observed that few proxies take a longer time to reveal the sentiment. To set out this, the study includes both levels and their lags while constructing PCA (Baker & Wurgler, 2006). The study performs a two-stage process in the construction of a sentiment index based on Baker and Wurgler’s (2006, 2007) methodology. Initially, the raw sentiment index is constructed using seven orthogonalized proxy sentiment indicators and their lags, which yield a total of 14-factor loadings. Then the study computes the correlation coefficient between the raw sentiment index and seven orthogonalized proxy sentiment indicators and their lags. In the end, for the construction of the final sentiment index, proxies having higher correlation coefficients were considered (Baker & Wurgler, 2006). The correlation coefficient between the raw sentiment index and the final stage sentiment index (INDex) is 0.915 which suggests that the risk of losing information by discarding seven implicit proxies is less. The first PC accounts for 38.5% of the variance which shows the following sentiment index: (1) $${\text{INDex}}_{\mathrm{t}}=-0.0{\text{15NIPO}}_{\mathrm{t}-1}+0.{\text{639ADR}}_{\mathrm{t}}+0.210{\text{PE}}_{\mathrm{t}-1}+0.{\text{727PB}}_{\mathrm{t}-1}+0.{\text{927RSI}}_{\mathrm{t}}+0.{\text{895MFI}}_{\mathrm{t}}-0.220{\text{TURN}}_{\mathrm{t}-1}$$(1)

4. Results

4.1. Unit root test

ADF unit root test with trend and intercept was performed to ensure the stationarity of the data series for a consistent estimator. Table 1 shows all the ADF values of the Nifty 500 and sectoral indices. The ADF unit root test of Nifty 500 shows a t stat of (−13.721) and a p-value of (0.000) at a level. Thus, it is ensured that all the other sectoral values are stationary at the level.

Table 1. ADF unit root test results for stationarity.

Variables	Nifty	Auto	FMCG	IT	Metal	PSU	INDex
ADF(t)	−13.752	−12.959	−16.071	−14.852	−12.742	−14.160	12.530
p-Value	0.000	0.000	0.000	0.000	0.000	0.000	0.000
Stationarity	YES	YES	YES	YES	YES	YES	YES

Source: Authors’ compilation. Note: Test exhibits stationarity at 1% significance.

4.2. Descriptive statistics of proxies and economic variables

Figure 1 depicts the movement of the market index – NSE Return (thick blue line) and sentiment index – INDEx (thin grey line). It is observed that the stock market index and sentiment index move in the same direction. However, more variation in sentiment over the stock return is observed when the index moves downward. Table 2 shows the descriptive summary of Nifty 500 and sectoral returns. The Nifty 500 shows a mean value of 0.051 of return which is a benchmark return in the market. The FMCG sector return account for the mean value of 0.056 which is relatively high followed by the Auto sector with 0.055 mean value. Whereas the PSU and IT sectors accounted for 0.021, and 0.010 mean values respectively. It indicates that both FMCG and Auto sectors offer a higher return than the benchmark index. This explains that the FMCG and Auto sectors contribute relatively higher returns than the benchmark index returns during the selected study period. In addition, the IT sector shows a 0.826 standard deviation (SD) which is relatively high, whereas the FMCG sector shows an SD value of 0.272. Overall, the FMCG sector seems to be the better sector in terms of its return and risk in the selected study period. From Table 2 it is observed that all series are negatively skewed, and all the kurtosis statistics are more than 3 indicating that the distribution of all sectoral return series is skewed and leptokurtic.

Figure 1. Trend movement of Market index (NSE Return) and sentiment index (INDex). Source: Authors’ compilation.

Table 2. Descriptive statistics for Nifty and sectorial returns for the period 2004–2021.

Variable	Mean	Median	Minimum	Maximum	Std. dev	Skewness	Kurtosis
NSERET	0.051	0.071	−1.800	1.491	0.340	−1.214	7.413
AUTO	0.055	0.083	−2.048	1.331	0.383	−1.097	5.370
FMCG	0.056	0.062	−1.104	0.779	0.271	−0.569	1.645
IT	0.010	0.083	−11.061	0.912	0.825	−11.249	148.190
METAL	0.040	0.040	−2.966	2.076	0.554	−0.805	4.843
PSU	0.021	0.013	−2.183	1.828	0.553	−0.067	1.139
INDex	0.000	0.037	−2.706	2.122	1.000	−0.280	−0.243

Note: This provides summary statistics of selected sectors’ return as well as sentiments index. Source: Authors’ compilation.

4.3. Relationship between sentiment index and index returns

The study using INDex examines the impact of IS on Nifty 500 and sectoral returns. To test the hypothesis, the OLS regression method is employed which arrived at interesting facts on the nexus between IS and SR. The results reveal that Durbin-Watson static values are close to 2, which confirms no autocorrelation. Table 3 depicts the Variation Inflation Factors (VIF), and the results show that all considered variable’s VIF values are less than 10 which confirms no collinearity in the data set.

Table 3. Variation inflation factors.

Sentiment proxies	VIF
ADR	1.410
PB	2.213
PE	2.226
TURN	1.979
RSI	4.338
MFI	3.519
NIPO	1.063

Source: Authors’ compilation.

To correct the heteroskedasticity and autocorrelation problems, Newey and West’s standard error method is used (Aggarwal & Mohanty, 2018). As Nifty 500, sectoral indices and INDex are stationary at level the OLS regression can be performed (Aggarwal & Mohanty, 2018). The regression equation is as follows: (2) $$R_t=\alpha +\beta \times \mathit{\text{sentimen}}{t}_t+\mathit{\text{contro}}{l}_t+{€}_t$$(2) where R_t is the return of the Nifty 500 and sectoral indices for time t, β is the regression coefficient, control includes control variables such as DY, T bill, FII, ER, and USEPU, and € is the error term.

Table 4 reports regression results. The result of the hypothesis in relation to INDex and Nifty 500 shows a positive coefficient (β = 0.198) at a 1% significance level. It reveals that IS significantly influences the Nifty 500 where the 1% change in sentiment leads to a 0.19% change in the return. Further, it shows that the sentiment index is a strong return predictor with a high adjusted R squared value of 0.645, which suggests that the regression model explains 64.5% of the variation in the stock market return. The result concerned with INDex and Auto shows a positive impact with a coefficient (β = 0.189) at a 1% level of significance. It implies that sentiment has a significant influence on the Auto sector. The 1% change in sentiment leads 0.18% change in the return. The results in connection with INDex and FMCG show a coefficient (β = 0.109) with a positive sign at a 1% level of significance. It implies that the 1% increase in sentiment leads to varying the return to the extent of 0.10%. As compared to other sectoral returns less sentiment effect is observed in the FMCG sector.

Table 4. Regression results.

Characteristics	Nifty	AUTO	FMCG	IT	Metal	PSU
Co-efficient	0.198***	0.189***	0.109***	0.200***	0.310***	0.261***
t-Stat	10.960	9.367	6.218	2.877	9.718	9.886
R^2	0.655	0.503	0.324	0.110	0.502	0.504
Adj R^2	0.645	0.488	0.304	0.084	0.488	0.489
DW statistics	2.335	2.001	2.123	2.122	2.029	2.282

Note: ***denotes significance level at 1%. Source: Authors’ compilation.

The results of the impact of INDex on IT sector show a positive influence. The findings show a positive significant effect with (β = 0.200) at a 1% level of significance. It indicates that the 1% increase in sentiment leads to varying the return to the extent of 0.20%. Similarly, the Metal sector shows that the sentiment is of paramount importance in influencing the return with a coefficient (β = 0.310) at a 1% level of significance. It is observed that INDex has a remarkable influence on the return of the Metal sector accounting for 0.31% of change in return. Furthermore, the findings of sentiment effect on the PSU sector also show the significant effect of sentiment with a coefficient (β = 0.261) at a 1% level of significance. It is noted that there is a significant sentiment effect on the returns of the PSU to the extent of 0.26%. While all coefficient values accounted for positive signs at a 1% level of significance, H1 is accepted due to the high sentiment effect on the SR of all indices. Overall, the adjusted R squared value is higher in all the regression results, it clearly shows that all considered variables can explain the variability in the SR.

5. Discussion

The present study aimed at examining the impact of IS on the SR of Nifty 500 and selected sectoral indices such as the Auto, Metal, IT, FMCG, and PSU. To test the framed hypothesis, the OLS regression was used. The findings of the study revealed that there exists a significant sentiment effect in the financial markets which causes changes in the stock market return. These findings are in tandem with (Aggarwal & Mohanty, 2018; Chakraborty & Subramaniam, 2020; Dash & Mahakud, 2012, 2013; Dash & Maitra, 2018; Pandey & Sehgal, 2019) who showed similar results in the emerging markets.

The rationale for the significant effect of INDex on all the sectoral indies is the active role of retail investors in the Indian equity market which accounts for 7.18%. This increasing participation of retail investors with insufficient professional knowledge and risk awareness in the equity markets is more likely to have irrational trading behavior patterns that lead to the exertion of a strong sentiment effect. Further, it is also observed that institutional investors dominate the Indian equity market (Badhani et al., 2023) and they are more sensitive to macro events, and their behavior guides the prevailing market sentiments. Since institutional investors play their role as ‘noise traders’ in the stock market, their sentiment is viewed as a major sentiment proxy (DeVault et al., 2019; Schmeling, 2007). Furthermore, Duxbury and Wang (2023) posit that retail as well as institutional investors cause inefficiency in the stock market.

It was found that the returns of Metal, PSU, IT, and Auto accounted for a high degree of sentiment effect. Similarly, the returns of the FMCG sector evidenced less sentiment effect as compared to these sectors. The noteworthy point here is that the FMCG has less sentiment effect and serves as a high-return offering sector with moderate risk followed by Auto and NSE 500 indices. The heterogeneity among these indices could be due to the different roles played by the investors in the markets. In addition, the strong sensitive effects of IS on sectoral return could be due to the changes in fund flow, and foreign activity (Aggarwal & Mohanty, 2018).

It is well evidenced that the findings depict sentiment can predict SR and exhibit a necessary association with the Nifty 500 and other sectorial indices. This observation is in line with the previous literature (Pyo & Kim, 2021; Wang et al., 2022). Further, study findings are in align with Brown and Clif (2005), Kumar and Lee (2006), and Baker and Wurgler (2007) in which it is observed that when investors are optimistic the stock will be sold at a premium and visa-versa. Particularly, when sentiment is anticipated to be high, the industries are attractive to optimists and speculators. It can be inferred that the higher the sentiment, the higher the return which leads to the overreaction of stock. Given the remarkable influence of IS in the country’s financial markets, one could understand the fluctuations of SR in the stock market. The results also highlight the symmetric causal behavior of IS on sectoral returns.

6. Conclusion

The study revealed interesting key facts about sentiment return relation in the Indian equity market. The study demonstrated that the INDex has a strong significant effect on Nifty 500 and selected sectorial indices return. The results also uncovered the fact that IS is of great importance in influencing the SR of selected indices. As IS is correlated with the index return, it is understood that the increase or decrease in SR results from a strong sentiment effect. Further, the results posit that the investors in the Indian equity market significantly affect the stock market performance and their irrationality continues to occur. As the sentiment effect is highly significant with a positive effect, it is understood that higher the sentiment, higher the return. Moreover, the study proves that sentiment is a major qualitative non-fundamental factor in driving the SR in the market.

The study findings contribute to the existing body of knowledge in several ways and have practical implications for stakeholders. The study results highlight the major antecedents of IS particularly indirect market proxy indicators in the Indian equity market. With a high sentiment-return relation in the market, retail and institutional investors need to take caution while building their portfolios. Further, the results can help policymakers to design stable policies to reduce the uncertainty in the market. Hence, the study provides new insights in connection with sentiment-return relation to major market participants and helps the researchers in understanding the crucial role of IS in the Indian equity market.

However, the paper is not without any limitations. The study considered a few sectors and proxies based on their data availability. Future research could include some other sectors such as health care, commodities, real estate, housing, and environmental, social, and governance in examining the sentiment effect on SR. Further, researchers are required to employ different proxy measures including social media measures with different frequencies. Furthermore, comparative studies can be done by considering various sectors in emerging markets. As the markets are extremely unpredictable due to investors’ emotional participation, research on sentiments needs to be continued for a better understanding of investors’ irrationality. Further studies are also encouraged to examine the sentiment effect in other forms of markets such as derivatives, bonds, and cryptocurrency markets.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Aditi N. Kamath

Aditi N Kamath is a research scholar in the Department of Commerce, Manipal Academy of Higher Education, Manipal. Her areas of interest include Finance and Accounting. Her current research focus is in the area of Behavioral Finance. She has published papers in Scopus and ABDC-listed journals.

Sandeep S. Shenoy

Dr. Sandeep S Shenoy is a Professor and Head at the Department of Commerce, Manipal Academy of Higher Education, Manipal. The author is specialized in the area of Finance and Management. His area of interest includes Finance, Sustainable Development, and Marketing. He has 21 publications in Scopus and ABDC-listed journals.

Abhilash Abhilash

Abhilash is a Senior Research Fellow at the Department of Commerce, Manipal Academy of Higher Education, Manipal. He has published papers in Scopus and ABDC-listed journals. His area of research includes Finance, Sustainability, and Corporate Governance. His current research focus is in the area of Green Bonds and Sustainability.

Subrahmanya Kumar N

Dr. Subrahmanya Kumar N is an Associate Professor at the Department of Commerce, Manipal Academy of Higher Education, Manipal. He has a keen interest in pursuing research on various aspects of Financial Management, Behavioral Finance, Management Accounting, and Cost Accounting. He has published papers in Scopus and ABDC-listed journals.