Journal of Islamic Monetary Economics and Finance.
Vol.
No.
, pp.
465 - 490
p-ISSN: 2460-6146, e-ISSN: 2460-6618 INFORMATION EFFICIENCY IN THE U.
AND SHARIAHCOMPLIANT STOCKS IN MALAYSIA DURING COVID-19
Ooi Kok Loang SEGi University.
Malaysia, kokloangooi94@hotmail.
ABSTRACT
This study examines the impact of analystsAo forecast on market liquidity and information efficiency in the U.
S .
and Malaysia .
merging Ae Shariah-compliant stock.
before and during COVID-19.
The results show that the analystsAo forecast is significant to the market liquidity in the pre-COVID period but its influence diminishes during the COVID-19.
Moreover, the impact of the analystsAo forecast is significant in the upper quantiles .
7 and 0.
9 quantile.
of the U.
S market and in the lower quantiles .
1 and 3 quantile.
of MalaysiaAos Islamic market.
Similarly, the buy-sell recommendations in the U.
S market and all variables forecasted are significant before COVID-19.
Both markets become inefficient during COVID-19, and analystsAo forecast is no longer correlated to information efficiency.
These results inform practitioners and investors to inspect the market conditions and investorAos behaviour under market stress such as COVID-19, which has disrupted the international financial markets.
Keywords: AnalystsAo forecast.
Market liquidity.
Information efficiency.
Investor behaviour.
COVID-19.
JEL classification: C23.
G4.
G14.
G15.
Article history:
Received : June 30, 2022 Revised : January 4, 2023 Accepted : August 31, 2023 Available online : September 29, 2023 https://doi.
org/10.
21098/jimf.
Information Efficiency in the U.
and Shariah-Compliant Stocks in Malaysia
During COVID-19
INTRODUCTION
The expert forecasts are a piece of information accessible by market participants.
The analystsAo forecasts of a stock performance are normally based on thorough financial analysis, and investors often respond to them.
In scholarly research, analyst forecasting is linked to market liquidity and information efficiency.
According to the literature on market liquidity, investorsAo sentiments and expert forecasts might impact liquidity (Abudy, 2.
The dissemination of new information to the market is facilitated by analyst projections, enabling investors to react accordingly (Blankespoor.
Miller & White, 2.
During a crisis, market liquidity is also more apparent (Galariotis.
Krokida & Spyrou, 2.
In selected markets, such as South Africa, the United Kingdom.
Germany.
Chile, the United States, and Malaysia, there are studies on the association between analyst forecasts and market liquidity in the normal period (Dang.
Doan.
Nguyen.
Tran & Vo.
Nevertheless, few studies evaluate the effect of expert forecasts on market liquidity during a crisis, particularly during COVID-19 pandemic.
Comparative research between established countries like the United States and developing markets like Malaysia is even scarce.
Many Shariah-compliant equities are traded on the Malaysian stock exchange.
Although other markets, such as Singapore and Egypt, also trade Shariah-compliant equities, the Malaysian market has a greater number of Islamic firms.
Moreover, the Malaysian market is dominated by Muslim investors, who may have diverse investment objectives (Barom, 2.
Ortmann.
Pelster & Wengerek .
demonstrate that amid market stress, investors respond to information differently.
Investors may increase their trading activity and sell shares to prevent investment loss due to panic trading.
Pandey & Kumari .
demonstrate that investors may exhibit irrational behaviour during crisis episodes.
Consequently, the behaviour of investors, particularly in reaction to the new information disclosed by analysts during the times of stress such as the COVID-19 outbreak, should be analysed.
Based on a few studies, investorsAo behaviour in established and developing markets does vary during market stress due to events such as COVID-19 pandemic.
Analyst forecasting serves as an information.
Analyst forecasting is related to information efficiency.
The Efficient Market Hypothesis (EMH) asserts that efficient markets should rapidly reflect all available market data on stock prices.
During COVID-19, the efficiency of several markets, including Saudi Arabia (Syed & Bajwa, 2.
Europe.
Malaysia, and the United States (Dias.
Teixeira.
Machova.
Pardal.
Horak, & Vochozka, 2.
, are thoroughly analysed.
However, no research addresses the association between analyst forecasting as a source of new information and the stock marketAos information efficiency.
This is because past research has often focused on the form of market efficiency rather than its drivers.
Additionally, though it is important, studies on variations in efficiency between established and developing markets pre- and during-COVID-19 are still lacking.
Most research evaluates the link between analyst forecasting, market liquidity, and information efficiency using the basic ordinary least squares (OLS) regression However, the OLS method does not control for unobserved variables in the regression, which might result in a biased conclusion.
In this situation, panel data and quantile regressions might be used as alternatives to the OLS.
Moreover, the quantile regression enables examination of the influence of analyst forecasts in Journal of Islamic Monetary Economics and Finance.
Vol.
Number 3, 2023 various quantiles of market liquidity since investors may react differently across Therefore, this study examines the impact of analyst forecasting on market liquidity and information efficiency before and during COVID-19 using panel data and quantile regressions.
It takes the U.
S market and the Malaysian Islamic markets as case studies.
The U.
S is the WorldAos largest stock market while Malaysia is one of the largest Islamic markets that trade Shariah-compliant stocks.
The comparison between the two thus will shed some light on the differences in the market efficiency between the developed and emerging (Malaysi.
markets, especially during the pandemic.
The rest of the paper is structured as follows.
The next section reviews related literature and develop hypotheses to be tested.
Then, section 3 presents the models and data.
This is followed the results in section 4.
Finally, section 5 provides conclusion of the paper.
II.
LITERATURE REVIEW AND HYPOTHESIS DEVELOPMENT
Analyst Forecasting.
Market Liquidity and Information Efficiency Analyst forecasting is the investment suggestion revealed by analysts through rigorous financial analyses (Firth.
Lin.
Liu & Xuan, 2.
Analyst forecasting is the predictions made by analysts who analyse the fundamentals and prospects of individual companies to assist investors in making wise investment decisions.
Analyst forecasting can affect market liquidity.
The study by Zyyiga.
Pincheira.
Walker & Turner .
shows that analyst forecasting with lower forecast errors can lead to higher market liquidity.
This is because the investors respond to the information released by analysts, and their trading behaviour is subsequently incorporated into stock prices.
Similarly.
Aouadi.
Arouri & Roubaud .
argue that information can affect market liquidity as investors make use of the information to make investment decisions.
They show that investors can be affected by new information from analysts in the market .
ee, for example.
Dang et al.
, 2019.
DeBoskey & Gillett, 2.
None of the studies looks at the impact of analyst forecasting on market liquidity in COVID-19, as investors can behave differently under stress and the market is disrupted during a pandemic (Loang & Ahmad, 2.
Numerous papers have been devoted to studying information efficiency (Lalwani & Meshram, 2020.
Dias et al.
, 2.
, but limited studies are treating analyst forecasting as the determinant of information efficiency.
Lalwani & Meshram .
argue that markets have become inefficient with the emergence of the pandemic because the information is delayed in being transmitted to the markets.
Besides.
Vasileiou .
shows that the U.
S market has become less efficient when investors are panicking and consequently selling off securities.
Nonetheless, no study examines the determinants of information efficiency, especially during the turbulent periods such as COVID-19.
The EMH states that markets shall reflect all private and public information, and efficient markets shall incorporate information into stock prices faster than less efficient markets.
From this perspective, the New York Stock Exchange (NYSE) is the WorldAos largest exchange and is expected to be more efficient than Information Efficiency in the U.
and Shariah-Compliant Stocks in Malaysia During COVID-19 Bursa Malaysia.
Furthermore, the Malaysian Islamic market lacks the same level of openness and accountability as the U.
Therefore, the Malaysia Islamic market shall be less efficient than the U.
S market.
Nevertheless, no study has compared the information efficiency of these two markets, especially during the COVID-19 pandemic.
Furthermore, most studies explore the impact of analyst forecasting using the ordinary least square (OLS) technique.
The OLS technique is not appropriate for pooled time series and cross-sectional data (Loang & Ahmad, 2.
Therefore, panel data regression is an alternative technique that accounts for unobserved variables in the regression and gives an in-depth look at the influence of analyst forecasting on market liquidity and information efficiency.
Furthermore, quantile regression, which measures the conditional median, can better address the impact of analyst forecasting on different quantiles of market liquidity.
Quantile regression can provide a more comprehensive result than the OLS method.
Hence, this study examines the impact of analyst forecasting on market liquidity and information efficiency before and during COVID-19 by using panel data and quantile regressions.
The following hypotheses are proposed:
H1.
Analyst forecasting is correlated to market liquidity before and after COVID-19.
H2.
Analyst forecasting is correlated to information efficiency before and after COVID-19.
H2.
MalaysiaAos Islamic market is less efficient than the U.
S market i.
ESTIMATED MODELS AND DATA
Market Liquidity In finance, liquidity is a complex concept with many measures.
A classical and conventional approach to measuring market liquidity is AmihudAos illiquidity measure (ILLIQ), as proposed by Amihud et al.
ILLIQ measures the magnitude of stock return at a given trading volume.
It captures the transaction cost per volume and considers the bid-ask spread as part of the return measurement.
The ILLIQ is estimated as:
Where, |Ri,.
is the absolute value of return on stock i on day d at period t.
VOLi,d,t is the daily volume in of stock i on day d and Ni is the number of trading days of stock i in period t.
A higher value of ILLIQ indicates that the stock is less liquid.
Nonetheless.
Lou & Shu .
argue that AmihudAos measure heavily relies on the trading volumes and hence fails to capture the price impact and lead to a biased In this context.
Bernales.
Cayyn & Verousis .
argue that the bid-ask spread is an alternative approach that can better examine market liquidity.
The bid-ask spread is simply the difference between the best bid and best ask prices that buyers and sellers are willing to accept in the market.
The relative bid-ask spread by using daily bids and asks is given as:
Journal of Islamic Monetary Economics and Finance.
Vol.
Number 3, 2023 Where, aski,t is the best ask price of stock i on day t and bidi,t is the best bid price of stock i on day t.
The lower bid-ask spread indicates higher liquidity of the stocks.
Another measurement of market liquidity is the turnover ratio.
As proposed in the study of Fan .
, the author argues that the turnover ratio captures the total shares traded in the market while not considering the costs such as price impact in the bid-ask spread.
The turnover ratio is estimated as :
Where.
Si is the total number of outstanding shares.
A higher value of turnover ratio shows that the stocks have higher liquidity.
Information Efficiency Information efficiency measures the accuracy of the information revealed by analysts in predicting the companiesAo future performance.
The information efficiency can be determined by the proportion of the information revealed by analysts compared to the actual stock performance (Hou.
Zhao & Yang, 2.
That is, .
One of the critical ratios forecasted by analysts is earning-per-share (EPS).
Analysts forecast EPS to predict the expected profit divided by the total number of outstanding shares.
It indicates a companyAos profitability, which directly impacts stock price and affects market response.
Therefore, the information revealed by analysts can be proxied by relative EPS, expressed as:
Where.
AFEi,t is the absolute value of forecast error of stock i at time t.
FEPSi,t is the forecast EPS of stock i at time t.
AEPSi,t is the actual EPS of stock i at time t.
Pi,t is the stock price of stock i at time t.
RFAi,t is the relative forecast accuracy of stock i at time t.
AFEmaxi,t is the maximum value of relative forecast accuracy of stock i at time t and AFEmini,t is the minimum value of relative forecast accuracy of stock i at time t.
Information Efficiency in the U.
and Shariah-Compliant Stocks in Malaysia During COVID-19 The second part of measuring information efficiency is determining the information revealed in stock performance.
The relative forecast accuracy can be transformed into relative information efficiency by capturing the impact of stock return synchronicity.
The relative information efficiency is calculated as follows:
Where.
i,t is the information efficiency of stock i at time t.
RIEi,t is the relative information efficiency of stock i at time t.
R2i,t is the stock return synchronicity of stock i at time t and IEmaxi,t and IEmini,t are the maximum and minimum values of information efficiency of stock i at time t.
AnalystsAo Forecasting Other than the EPS forecasted by analysts (Eq.
, analysts also predict other financial ratios such as return on assets (ROA), return on equity (ROE), book value per share (BVPS) and earnings before interest, taxes, depreciation and amortisation (EBITDA).
ROA and ROE allow the analysts to evaluate the performance of the management in utilising total assets and equities as the companyAos resources to generate profit.
The relative ROA can be expressed as:
Where.
RROAi,t is the relative forecast ROA of stock i at time t.
AROAmaxi,t and AROAmini,t are the maximum and minimum of forecast ROA of stock i at time t and AROAi,t is the actual ROA of stock i at time t.
The relative ROE is given as:
Where.
RROEi,t is the relative forecast ROE of stock i at time t.
AROAmaxi,t and AROAmini,t are the maximum and minimum of absolute forecast ROE of stock i at time t and AROAi,t is the actual ROE of stock i at time t.
Furthermore.
BVPS reveals the companyAos net asset value on a per-share basis to evaluate whether a stock is The relative BVPS guides investors in stock selection as follow:
Where.
RBVPSi,t is the relative forecast BVPS of stock i at time t.
ABVPSmaxi,t and ABVPSmini,t are the maximum and minimum of forecast BVPS of stock i at time t and ABVPSi,t is the actual BVPS of stock i at time t.
Moreover.
EBITDA is an alternative measurement to net profit without considering the cost of capital investments.
Journal of Islamic Monetary Economics and Finance.
Vol.
Number 3, 2023 is a simple measurement that analysts used as a metric of companiesAo profitability.
The relative EBITDA is expressed as:
Besides the financial ratios, analysts also provide buy-sell recommendations to investors to inform investors of suggested trading action.
In this study, buysell recommendations are categorised into 5 different groups with 1 denoting a AuStrong BuyAy, 2 is AuBuyAy, 3 is AuHoldAy, 4 is AuSellAy and 5 is AuStrong SellAy, as in Loang & Ahmad .
Control Variables Company-specific factors such as leverage, company size, volatility, quick ratio and price/earnings to growth (PEG) ratio are included as control variables given that they have been found to be correlated to market liquidity and stock return.
Leverage is measured by total liabilities on total assets.
Company size is proxied by market capitalisation.
Volatility is calculated by realised volatility.
The quick ratio measures the companyAos ability to meet short-term liquidity using current assets minus inventory divided by current liabilities.
PEG ratio is a stock valuation measurement using the P/E ratio divided by the growth rate of a company.
Panel Data and Quantile Regressions Panel data regression combines cross-sectional and time-series data and allows for unobserved variables by specifying the individual-specific effect as either fixed or It provides greater explanatory power compared to the OLS method.
The panel data regressions for market liquidity and information efficiency are written respectively written as:
Where.
Recomi,t is the average buy-sell recommendations of stock i at time t.
Levi,t is the leverage ratio of stock i at time t.
FSi,t is the company size of stock i at time t.
Volai,t is the realised volatility of stock i at time t.
Qi,t is the quick ratio of stock i Information Efficiency in the U.
and Shariah-Compliant Stocks in Malaysia During COVID-19 at time t.
PEGi,t is the PEG ratio of stock i at time t, and all other variables are as defined earlier.
Furthermore, quantile regression allows this study to examine the impact of analystsAo forecasting in different quantiles of market liquidity.
Unlike the OLS, the quantile regression measures conditional median rather than condition mean.
The quantile regression is given as:
Where.
Yi is the dependent variable, xi is the vector of the independent variable and is the vector of coefficient.
By minimising weighted deviations from the conditional quantile, the parameter vector of the E-th quantile of the conditional distribution is expressed as (Jiang.
Zhang & Sun, 2.
The quantile loss function is written as:
When, ui= yi-xiAo , the Equation .
can be defined as:
Equation .
shows that the quantile regression estimates can be measured when the weighted sum of the absolute errors are minimised.
The weights are dependent on the quantile values.
Therefore, the quantile regression can be expressed as .
In the analysis, the quantile regression is employed for market liquidity model.
Data The data span from 1-Jan-2014 to 31-Oct-2021 and only the stocks listed in NYSE and Bursa Malaysia were selected.
The last seven years of data analysis allow this study to comprehensively evaluate stock market performance (Chatzis.
Siakoulis.
Journal of Islamic Monetary Economics and Finance.
Vol.
Number 3, 2023 Petropoulos.
Stavroulakis & Vlachogiannakis, 2.
The U.
S has the WorldAos largest and most developed stock market, and Malaysia is an emerging market.
For comparison analysis, the data for pre-COVID-19 from 1-Jan-2014 to 1-Dec2019 while the data for COVID-19 ranges from 1-Jan-2020 to 31-Oct-2021 are used.
Quarterly bank-specific data are also gathered.
Other securities, such as funds and warrants, are excluded.
All stocks shall be listed before Jan-2014 and maintain the listed status until Oct-2021.
The sample size is 1287 from NYSE and 527 from Bursa Malaysia.
As regards to shariahcompliant stocks in Bursa Malaysia, only those stocks that remain listed as of 31 December 2021 are chosen.
Shariah-compliant stocks safeguard stakeholdersAo interests, prohibiting riba, gharar, suspicious transactions and gambling.
All data are collected from the Standard and PoorAos (S&P) Capital I.
Database.
IV.
EMPIRICAL RESULTS AND ANALYSIS
Descriptive statistics Table 1 provides descriptive statistics of the variables for both NYSE and Bursa Malaysia sample stocks.
The NYSE has all positive mean values of analystsAo forecasting, while Malaysia has all negative mean values of analystsAo forecasting.
It shows that analysts in NYSE tend to provide higher forecast values than actual stock performance.
On the other hand, analysts in the Malaysian Islamic market predict decline in stock values in the near future.
This difference can be caused by the information efficiency between emerging and developed markets.
Table 1.
Variables of NYSE and Malaysia
Mean
NYSE
Recommendation EPS
ROE
ROA
BVPS
EBITDA
Malaysia
Recommendation EPS
ROE
ROA
BVPS
EBITDA
Median Maximum Minimum Std.
Dev Skewness Kurtosis Information Efficiency in the U.
and Shariah-Compliant Stocks in Malaysia During COVID-19 Estimate of AnalystsAo Forecasting and Market Liquidity in The US Market liquidity is represented by three different measures - AmihudAos ILLIQ, bid-ask spread and turnover ratio.
The analystsAo forecasting is proxied by buysell recommendation.
EPS.
ROE.
ROA.
BVPS and EBITDA as revealed in analyst Panel data regression is used to determine the impact of analystsAo forecasting on market liquidity.
The Hausman test evaluates whether a random or fixed model is suitable.
According to Loang & Ahmad .
, unobserved variables can have correlations with observable variables in a fixed-effects model, where the estimates would be consistent.
On the other hand, random effect models presume that individual characteristics are unrelated to the dependent variable.
Table 2.
Impact of Analyst Forecasting and Company Information on Market Liquidity of NYSE before and during COVID-19
Panel Data
Constant
Before COVID-19
Bid-ask
ILLIQ
TURN
Spread
FixedFixedRandomEffect
Effect
Effect
During COVID-19
Bid-ask
ILLIQ
TURN
Spread FixedFixedRandomEffect Effect Effect .
001***
079***
085***
002***
(-4.
163***
599***
(-7.
776***
(-0.
(-2.
373***
004***
(-2.
008***
(-0.
002***
(-1.
(-1.
116***
092***
(-2.
178***
(-4.
071***
(-1.
(-0.
(-0.
(-0.
(-0.
(-0.
(-0.
(-1.
209***
(-16.
002***
047***
(-0.
(-1.
000***
003***
(-17.
000***
(-0.
(-0.
038***
(-15.
002***
006***
(-11.
(-1.
(-1.
142***
(-2.
008***
(-9.
(-0.
002***
001***
000***
(-0.
010***
(-10.
(-0.
Analyst Forecasting Recommendation EPS
ROE
ROA
BVPS
EBITDA
Company Information Leverage Company Size
Volatility Quick
PEG
Journal of Islamic Monetary Economics and Finance.
Vol.
Number 3, 2023 Table 2.
Impact of Analyst Forecasting and Company Information on Market Liquidity of NYSE before and during COVID-19 (Continue.
Panel Data
Specification Tests Adjusted R2
Hausman Test
Time Effect
Modified Wald
Breush-Pagan Pesaran Scaled Pesaran CD
Durbin-Watson
Before COVID-19
Bid-ask
ILLIQ
TURN
Spread
FixedFixedRandomEffect
Effect
Effect
During COVID-19
Bid-ask
ILLIQ
TURN
Spread FixedFixedRandomEffect Effect Effect Yes Yes Yes Yes Yes Yes Note: ***, ** and * shows that significant level of 1%, 5% and 10%.
Table 3.
Quantile Regression of NYSE ILLIQ.
TURN and Bid-Ask Spread before and during COVID-19 Analyst Forecasting Recom.
EPS
ROE
ROA
BVPS
EBITDA
Recom_COVID EPS_COVID ROE_COVID ROA_COVID BVPS_COVID EBITDA_COVID Company Information Leverage Company Size
Volatility Quick
PEG
Leverage_COVID Company Size_COVID Volatility_COVID Quick_COVID PEG_COVID NYSE ILLIQ
035***
010***
133***
024***
046***
013***
274***
053***
046***
128***
034***
486***
103***
040***
140***
294***
084***
523***
259***
172***
471***
022***
220***
576***
924***
001***
015***
023***
008***
019***
001***
000***
000***
017***
001***
022***
002***
001***
018***
001***
001***
024***
004***
001***
018***
000***
002***
003***
030***
000***
010***
002***
016***
000***
004***
005***
025***
039***
000***
003***
011***
015***
012***
Information Efficiency in the U.
and Shariah-Compliant Stocks in Malaysia During COVID-19 Table 3.
Quantile Regression of NYSE ILLIQ.
TURN and Bid-Ask Spread before and during COVID-19 (Continue.
Analyst Forecasting Recom.
EPS
ROE
ROA
BVPS
EBITDA
Recom_COVID EPS_COVID ROE_COVID ROA_COVID BVPS_COVID EBITDA_COVID Company Information Leverage Company Size
Volatility Quick
PEG
Leverage_COVID Company Size_COVID Volatility_COVID Quick_COVID PEG_COVID 001*** 001*** 003*** 002*** 001*** 001*** 003*** 001*** 002*** 001*** 001*** 002*** 003*** 002*** 001*** 001*** 001*** 003*** 000*** 000*** 000*** 000*** 000*** 001*** 001*** 003*** 004*** 000*** 000***
000***
000***
000***
000***
000***
000***
000***
000***
000***
000***
000***
000***
000***
000***
000***
000***
001***
000***
000***
001***
000***
000***
000***
001***
000***
000***
001***
000***
000***
002***
000***
000***
000***
000***
000***
000***
041***
002***
008***
187***
203***
418***
104***
008***
003***
Analyst Forecasting Recom.
EPS
ROE
ROA
BVPS
EBITDA
Recom_COVID EPS_COVID ROE_COVID ROA_COVID BVPS_COVID EBITDA_COVID NYSE TURN
098***
NYSE Bid-ask Spread
181***
112***
151***
253***
100***
103***
137***
Journal of Islamic Monetary Economics and Finance.
Vol.
Number 3, 2023 Table 3.
Quantile Regression of NYSE ILLIQ.
TURN and Bid-Ask Spread before and during COVID-19 (Continue.
Company Information Leverage Company Size
Volatility Quick
PEG
Leverage_COVID Company Size_COVID Volatility_COVID Quick_COVID PEG_COVID 054*** 012*** 004*** 032*** 014*** 001*** NYSE Bid-ask Spread
057***
011***
000***
025***
000***
011***
049***
000***
003***
032***
000***
007***
033***
000***
001***
000***
036***
000***
006***
034***
000***
004***
000***
035***
007***
005***
Note: ***, ** and * shows that significant level of 1%, 5% and 10%.
Table 2 outlines the impact of company information and analyst forecasting on NYSE market liquidity before and during COVID-19.
A fixed effect model of panel data regression is adopted as the Hausman test is found to be significant except for the bid-ask spread, which uses the random effect model.
The result shows that all the variables of analyst forecasting are significantly related to market liquidity (ILLIQ.
TURN and bid-ask sprea.
Surprisingly, these variables of analyst forecasting, i.
buy-sell recommendation.
EPS.
ROE.
ROA.
BVPS and EBITDA, turn insignificant during COVID-19.
As for the control variables, leverage, company size and volatility are discovered to be significant to market liquidity before and during COVID-19.
There is no empirical evidence to indicate the impact of the quick and PEG ratios on market liquidity.
According to Table 2, the impact of company information is not affected by the emergence of COVID-19 in the NYSE.
Furthermore, the modified Wald test detects the existence of group-wise heteroscedasticity in a regression model.
The modified Wald test is insignificant, as shown in Table 2, and fails to reject the null hypothesis in claiming that the pane data regression is homoscedastic.
The alternative approaches to detecting heteroscedasticity are the BreushPagan test.
Pesaran Scaled test and the Pesaran CD test.
The Breusch-Pagan test determines whether the variance of regression errors is affected by the values of the independent variable in the regression.
Nonetheless, the Breusch-Pagan Test is ineffective for determining sample size with a large N.
Therefore.
Pesaran.
Schuermann, & Weiner .
present Pesaran Scaled and Pesaran CD as the standardised versions to address the Breusch-Pagan TestAos limitations.
Table 2 reveals that the Breush-Pagan.
Pesaran Scaled, and Pesaran CD tests are insignificant, indicating that the panel data regression is homoscedastic.
For robustness, quantile regression is adopted to examine the impact of analyst forecasting on different quantiles of market liquidity.
Table 3 shows the quantile Information Efficiency in the U.
and Shariah-Compliant Stocks in Malaysia During COVID-19 regression of NYSE ILLIQ.
TURN and bid-ask spread at the quantile of 0.
1, 0.
3, 0.
7 and 0.
The impact of analyst forecasting tends to appear in the upper quantile of market liquidity at 0.
7 and 0.
Figure 1.
, 1.
, and 1.
summarise the results of quantile regression for the NYSE ILLIQ.
TURN and bid-ask spread before and during COVID-19.
NYSE ILLIQ
Figure 1.
Quantile Regression of NYSE ILLIQ
NYSE TURN
Figure 1.
Quantile regression of NYSE TURN Journal of Islamic Monetary Economics and Finance.
Vol.
Number 3, 2023
NYSE Bid-ask spread
Recommendation EPS
ROE
ROA
BVPS
EBITDA
Leverage Mark.
Cap
Volatility Quick
PEG
Recommendation_COVID EPS_COVID ROE_COVID ROA_COVID BVPS_COVID EBITDA_COVID Leverage_COVID Mark.
Cap_COVID Volatility_COVID Quick_COVID PEG_COVID Figure 1.
Quantile Regression of NYSE Bid-Ask Spread Estimate of Analysts Forecasting and Market Liquidity in Malaysia Table 4 outlines the impact of company information and analyst forecasting on market liquidity in the Malaysian Islamic market before and during COVID-19.
The fixed effect model is selected for the panel data regression as the results of the Hausman test show that the null hypothesis is rejected with p-values less than Surprisingly, the overall result is similar to the NYSE.
Buy-sell recommendation.
EPS.
ROE.
ROA and EBITDA are significant to market liquidity at the significant level of 10%, 5% and 1% before COVID-19.
Nonetheless.
BVPS is found to be With the arrival of COVID-19, the impact of analyst forecasting has diminished, and all the variables of analyst forecasting are insignificant.
For company information, leverage, company size, volatility and quick ratio are found to be significant in different market liquidity measurements.
The explanatory power of these variables is not affected during COVID-19.
Besides, the PEG ratio is the only variable of the company information that is not significant.
Information Efficiency in the U.
and Shariah-Compliant Stocks in Malaysia During COVID-19 Table 4.
Impact of Analyst Forecasting and Company Information on Market Liquidity of Malaysia Before and during COVID-19
Panel Data
Constant
Before COVID-19
Bid-ask
ILLIQ
TURN
Spread
FixedFixedRandomEffect
Effect
Effect
During COVID-19
Bid-ask
ILLIQ
TURN
Spread
FixedFixedRandomEffect
Effect
Effect
956***
(-20.
006***
(-13.
079***
(-0.
012***
(-11.
498***
(-3.
162***
(-3.
050***
(-1.
(-1.
000***
001***
(-1.
000***
(-1.
116***
092***
(-2.
(-4.
071***
(-0.
(-0.
(-0.
(-0.
(-0.
(-0.
(-0.
(-0.
(-1.
(-0.
(-0.
(-0.
003***
(-12.
007***
(-6.
003***
025***
(-0.
004***
008***
(-5.
000***
(-1.
(-0.
038***
(-15.
002***
006***
(-11.
(-1.
(-1.
(-1.
(-0.
002***
002***
(-3.
000***
(-0.
(-0.
023***
(-0.
(-0.
Yes
Yes
Yes
Yes
Yes
Yes
Analyst Forecasting Recommendation EPS
ROE
ROA
BVPS
EBITDA
Company Information Leverage Company Size
Volatility Quick
PEG
Specification Tests Adjusted R2 Hausman Test Time Effect Modified Wald Breush-Pagan Pesaran Scaled Pesaran CD Durbin-Watson Note: ***, ** and * shows that significant level of 1%, 5% and 10%.
Journal of Islamic Monetary Economics and Finance.
Vol.
Number 3, 2023 Table 5.
Quantile Regression of Malaysia ILLIQ.
TURN and Bid-Ask Spread before and during COVID-19 Analyst Forecasting Recom.
EPS
ROE
ROA
BVPS
EBITDA
Recom_COVID EPS_COVID ROE_COVID ROA_COVID BVPS_COVID EBITDA_COVID Company Information Leverage Company Size
Volatility Quick
PEG
Leverage_COVID Company Size_COVID Volatility_COVID Quick_COVID PEG_COVID 000*** 000*** 000*** 032*** 001*** 000*** 051*** 244*** 226*** 015*** 009*** 000*** 000*** 014*** 012*** 028*** 000*** 000*** 000*** 012*** 013*** 034*** 000*** 001*** 008*** 010*** 044*** 000*** 000***
001***
006***
000***
041***
000***
000***
001***
002***
001***
003***s Analyst Forecasting Recom.
EPS
ROE
ROA
BVPS
EBITDA
Recom_COVID EPS_COVID ROE_COVID ROA_COVID BVPS_COVID EBITDA_COVID Malaysia ILLIQ
001***
001***
002***
003***
001***
001***
006***
000***
Malaysia TURN
002***
001***
002***
001***
004***
001***
000***
000***
002***
002***
001***
Information Efficiency in the U.
and Shariah-Compliant Stocks in Malaysia During COVID-19 Table 5.
Quantile Regression of Malaysia ILLIQ.
TURN and Bid-Ask Spread before and during COVID-19 (Continue.
Company Information Leverage Company Size
Volatility Quick
PEG
Leverage_COVID Company Size_COVID Volatility_COVID Quick_COVID PEG_COVID 000*** 000*** 000*** Analyst Forecasting Recom.
EPS
ROE
ROA
BVPS
EBITDA
Recom_COVID EPS_COVID ROE_COVID ROA_COVID BVPS_COVID EBITDA_COVID Company Information Leverage Company Size
Volatility Quick
PEG
Leverage_COVID Company Size_COVID Volatility_COVID Quick_COVID PEG_COVID Malaysia TURN
000***
000***
000***
000***
000***
000***
000***
000***
000***
000***
001***
001***
Malaysia Bid-ask Spread
000***
000***
001***
000***
000***
001***
001***
000***
000***
001***
000***
001***
000***
000***
000***
019***
002***
000***
057***
000***
004***
001***
000***
001***
000***
005***
005***
001***
014***
040***
035***
001***
043***
Note: ***, ** and * shows that significant level of 1%, 5% and 10%.
Journal of Islamic Monetary Economics and Finance.
Vol.
Number 3, 2023 Shanghai ILLIQ Figure 2.
Quantile Regression of Malaysia ILLIQ Shanghai TURN Figure 2.
Quantile regression of Malaysia TURN Information Efficiency in the U.
and Shariah-Compliant Stocks in Malaysia
During COVID-19
Shanghai Bid-ask spread
Recommendation ROE
BVPS
Leverage Volatility PEG
EPS_COVID
ROA_COVID
EBITDA_COVID
Mark.
Cap_COVID Quick_COVID EPS
ROA
EBITDA
Mark.
Cap Quick Recommendation_COVID ROE_COVID BVPS_COVID Leverage_COVID Volatility_COVID PEG_COVID Figure 2.
Quantile Regression of Malaysia Bid-Ask Spread The results for the NYSE and Malaysia show that the impact of analyst forecasting is no longer significant with the emergence of COVID-19.
It is inconsistent with the study of Bilinski .
in which the author argues that investors react strongly to buying and selling shares based on the revision of analyst forecasts and recommendations during COVID-19.
The author shows that investors value analyst information in investment decision-making.
Nonetheless, the result of this study indicates differently by showing no relation between analyst forecasting and market liquidity during the COVID-19.
One reason is that analysts tend to generate prediction errors during uncertain The unanticipated introduction of COVID-19 is a chaos to the market.
There are a lot of uncertainties in the information which affect the quality of the information used by analysts to come up with their forecast.
It is impossible for experts to accurately predict market performance.
There must be a delay in the release of fresh market information.
Therefore, investors disregard expert predictions to As a result of unforeseen occurrences, revisions of recommendations may also include several forecasting mistakes.
In this situation, expert forecasts fail to accurately anticipate stock performance.
Therefore, there is no relation between expert forecasts and market liquidity.
Journal of Islamic Monetary Economics and Finance.
Vol.
Number 3, 2023 Estimate of Information Efficiency in NYSE and Malaysia The second objective of this study is to examine the impact of company information and analyst forecasting on information efficiency.
As shown in Equation 4, information efficiency is measured by the difference between the information revealed by analysts compared to the actual companyAos performance.
The fixed effect model is chosen for the panel data regression as the Hausman tests have p-values less than 0.
The results of the modified Wald test.
Breush-Pagan.
Pesaran Scaled and Pesaran CD indicate that the panel data regression contains heteroscedasticity with p-values less than 0.
Therefore, this study adopts panel-corrected standard error (PCSE) to rectify heteroscedasticity for pre-COVID-19.
Table 6 summarises the results of the relation between company-specific information, analyst forecasting and information efficiency in the NYSE and Malaysia before and during COVID-19.
For the NYSE, the result shows that buysell recommendation is the only variable of analyst forecasting with a significant impact on information efficiency before COVID-19.
The other variables.
ROE.
ROA.
BVPS and EBITDA, are insignificant.
Nonetheless, all variables of analyst forecasting are found to be significant to information efficiency in Malaysia before COVID-19 at the significant level of 1%.
With the emergence of COVID-19, none of the analyst predicting factors are important to the information efficiency of the NYSE and Malaysia.
It demonstrates that the information disclosed by analysts is not reflected in the stock prices during COVID-19.
Both markets are inefficient during the COVID-19.
During COVID-19, company information is also insignificant on the Malaysian Islamic market.
One of the reasons is that the markets are not as efficient as in normal times during the pandemic.
This is because COVID-19 has disrupted markets in response to analyst forecasts, and the amended information disclosed by analysts is not reflected in stock prices.
Investors respond differently to market volatility (Economou.
Hassapis & Philippas, 2.
The outcome is consistent with VasileiouAos .
examination of market efficiency during COVID-19.
contends that investorsAo anxiety has rendered the U.
market inefficient, causing them to act irrationally when selling assets.
Information Efficiency in the U.
and Shariah-Compliant Stocks in Malaysia During COVID-19 Table 6.
Correlation between Company-Specific Information.
Analyst Forecasting and Information Efficiency in NYSE and Malaysia
NYSE
Before During COVID-19
COVID-19
Fixed-Effect
Fixed-Effect
(PCSE)
Constant
Malaysia
Before During COVID-19
COVID-19
Fixed-Effect
Fixed-Effect
(PCSE)
(-0.
113***
(-6.
155***
196***
(-3.
(-1.
(-1.
(-0.
(-1.
(-0.
(-2.
153***
007***
(-1.
060***
049***
(-11.
019***
(-3.
(-1.
(-1.
025***
004***
(-4.
007***
(-0.
(-0.
025***
(-7.
003***
(-5.
012***
(-1.
(-1.
(-2.
(-2.
Yes
Yes
Yes
Yes
Analyst Forecasting Recommendation ROE
ROA
BVPS
EBITDA
Company Information Leverage Company Size
Volatility Quick
PEG
Specification Tests Adjusted R2 Hausman Test Time Effect Modified Wald Breush-Pagan Pesaran Scaled Pesaran CD Durbin-Watson Note: ***, ** and * shows that significant level of 1%, 5% and 10%.
Journal of Islamic Monetary Economics and Finance.
Vol.
Number 3, 2023 Robustness Checks For robustness test, this study adopts the granger causality test to examine the potential causality between variables.
The result presented in table 7 indicates that all independent variables are causally linked to the dependent variables, which are the market liquidity and information efficiency.
Hence, the empirical evidence suggests that analystsAo forecasting Granger causes market liquidity and efficiency in US and Malaysia.
Table 7.
Robustness Model Variable REC does not Granger Cause LIQ LIQ does not Granger Cause REC ROE does not Granger Cause LIQ LIQ does not Granger Cause ROE ROA does not Granger Cause LIQ LIQ does not Granger Cause ROA BVPS does not Granger Cause LIQ LIQ does not Granger Cause BVPS EBITDA does not Granger Cause LIQ LIQ does not Granger Cause EBITDA LEV does not Granger Cause IE IE does not Granger Cause LEV SIZE does not Granger Cause IE IE does not Granger Cause SIZE VOL does not Granger Cause IE IE does not Granger Cause VOL QUI does not Granger Cause IE IE does not Granger Cause QUI PEG does not Granger Cause IE IE does not Granger Cause PEG Malaysia P-Value P-Value CONCLUSION This study examines the impact of analyst forecasting on market liquidity and information efficiency in the U.
S (NYSE Ae developed marke.
and Malaysia (Bursa Malaysia - emerging market of Shariah-compliant stock.
before and during the COVID-19 pandemic.
The data cover the period from 1-Jan-2014 for pre-COVID-19 and 1-Jan-2020 to 31-Oct-2021 for COVID-19.
Panel data and quantile regressions are adopted in the analysis.
Market liquidity is represented by AmihudAos ILLIQ.
TURN and bid-ask spread.
The information revealed by analysts is proxied by buy-sell recommendation.
EPS forecast.
ROE forecast.
ROA forecast.
BVPS forecast and EBITDA forecast.
Various company-specific characteristics are also included as control variables.
The results of this study show that all variables of analyst forecasting are significant to the U.
All variables except BVPS are significant to MalaysiaAos Islamic Information Efficiency in the U.
and Shariah-Compliant Stocks in Malaysia During COVID-19 marketAos liquidity.
Nonetheless, the impact of analyst forecasting has diminished with the emergence of COVID-19.
This is because COVID-19 is disastrous to the markets, and analysts faced a lot of uncertainties to provide accurate information to investors.
Hence, investors may not rely on analyst information to trade and no significant relationship between analyst forecasting and market liquidity during the COVID-19 pandemic.
For the relationship between analyst forecasting and information efficiency, the buy-sell recommendation is the only variable of analyst forecasting that is found to be related to information efficiency in the U.
S before COVID-19.
All variables of analyst forecasting are significant to the information efficiency of Malaysia.
Surprisingly, analyst forecasting is insignificant with the arrival of COVID-19.
One of the reasons is that the markets are inefficient during the pandemic because analysts have yet to respond to unexpected events, and new information can be delayed to incorporate into stock prices.
The results of this study contribute to the academic research of behavioural finance, academic scholars and investors in understanding the impact of analyst forecasting on market liquidity and information efficiency, especially during the pandemic.
This study shows that the influence of analyst information has diminished with COVID-19, as investors can react differently under market The results suggest that analysts release new information promptly and Otherwise, the investors may not rely on their recommendations to make investment decisions.
Policymakers and regulators shall be aware of the determinants affecting market liquidity and information efficiency, which signals a financial crisis.
REFERENCES