Understanding stock reactions to quarterly financial results announcements

• Aim: Several macro economic and company-level factors affect stock price movements of a company. The aim of this article is to identify them using literature review method.
• Method: At least 30 studies conducted in the past 5 years must be considered.
• Analysis/ presentation: The factors must be presented in separate sections with detailed explanation of how, why, and when they affect stock prices. Specify the most popular method to estimate/ measure the impact of the factor. 
• Conclusion: At the end of the article, a conceptual diagram must be presented displaying the factors with classification into 'organisational' and 'macroeconomic' factors. 

• Aim: To review studies on impact of results announcement on stock prices
• Methodology: Perform a systematic review of at least 25 studies published in the last 5 years (post 2018) on the impact of financial results announcement on stock prices. There is no restriction on the geographical location of the studies. Use PRISMA process to identify the relevant studies. Here is a sample systematic review.
• Analysis/ presentation: Systematic review must be presented in the form of a table showing Author (Year), Aim of the study, Methodology (event window & analysis method), Variables, and Findings. After the table, the DISCUSSION section must summarise the findings. All the reference papers must be submitted along with the article in a zip folder.

Aim: To review studies on impact of trade volume on stock prices

Methodology: This article includes a systematic review of at least 10 studies published in the last 5 years (post 2018) on the impact of trade volume on stock prices. There is no restriction on the geographical location of the studies. Use PRISMA process to identify the relevant studies. Here is a sample systematic review.
Analysis/ presentation: Systematic review must be presented in the form of a table showing Author (Year), Aim of the study, Methodology (event window & analysis method), Variables, and Findings. After the table, the DISCUSSION section summarising the findings must also be presented. All the reference papers must be submitted along with the article in a zip folder.

• Aim: To review studies on impact of market performance on stock prices
• Methodology: This article includes a systematic review of at least 15 studies published in the last 5 years (post 2018) on the impact of market performance (i.e. stock market like NSE, BSE, NASDAQ)on stock prices. There is no restriction on the geographical location of the studies. Use PRISMA process to identify the relevant studies. Here is a sample systematic review.
• Analysis/ presentation: Systematic review must be presented in the form of a table showing Author (Year), Aim of the study, Methodology (event window & analysis method), Variables, and Findings. The DISCUSSION section must summarise the findings. All the reference papers used must be submitted along with the article. in a zip folder

• Aim: To review studies on impact of TYPE OF ANNOUNCEMENT (i.e. higher or lower than previous financial period) on stock prices
• Methodology: This article includes a systematic review of at least 15 studies published in the last 5 years (post 2018) on the impact of market performance (i.e. stock market like NSE, BSE, NASDAQ)on stock prices. There is no restriction on the geographical location of the studies. Use PRISMA process to identify the relevant studies. Here is a sample systematic review.
• Analysis/ presentation: Systematic review must be presented in the form of a table showing Author (Year), Aim of the study, Methodology (event window & analysis method), Variables, and Findings. The DISCUSSION section must summarise the findings. All the reference papers used must be submitted along with the article. in a zip folder

This milestone is in continuation of the milestone titled 'Variables and methodology for testing the impact of results announcement on share prices'. 

Aim: In this milestone we will describe the first step of data analysis process of Objective 1 (i.e. to assess the impact of results announcement on share prices) of Goal 2 (i.e. to create a prediction model for stock price movement post quarterly result announcement).

Calculation: this step involves calculating stock returns for the event window, i.e. T-2, T-1, T, T+1, T+2. 

Dataset: Stock returns is stock closing price minus opening price. This value is essential for our regression model which will yield the intercept and slope coefficient.

Structure: of the article will be as follows.

1. Introduction to the previous article (methodology of the study, especially dataset)
2. calculation process 
   • importing the data in STATA
   • process of calculating returns 
3. results representation 
   • trend shown in results sector wise (graph + explanation)
4. summary- explain the next step (regression for estimating the slope coefficient and intercept)

Article can be between 800 and 1000 words.

Additional info about the methodology: The article will include collection of opening and closing price of stock from FY 2018-2019 to FY 2022-2023 for each quarter announcement data and 2 days prior to it and 2 days post announcement. So, if our announcement date for June 2022 is 10 July 2022 then data will be collected for 2 time prior and 2 days post. Remember if the stock market is not functioning on any day then data will be taken for another time when stock market is open. For example if it is not open on 8th July 2022 then we will take data of 7th July 2022. Listing out all companies in this sector and then collecting data for closing and opening price of each company for this time period and also of that sector indices for closing and opening price. The analysis will be done using formula return = closing - opening price using STATA

This milestone is in continuation of the milestone titled 'Stock returns analysis of BSE listed companies'. 

Aim: To evaluate the BSE performance of the selected stocks for the FY 2017-2018 using time series regression analysis

Purpose: The overall purpose of this goal is to predict stock prices after a company's quarterly results are announced. But there are external factors too which affect stock prices. In this milestone we will test if there is any impact of market return, i.e. overall SECTORAL performance on a comapny's stock return. For example, Wockhardt's stock performance may be affected by the overall pharmaceutical sector's performance.  For this we will use time series regression test to calculate the slope coefficient (coefficient associated with the independent variable, ie sectoral market return) and intercept (coefficient associated with constant).

Data: Collect data from the BSE website for the following for the time period 1st April 2017 to 31st March 2018:

1. selected companies (select 'Equity' option in the link. Check the previous articles for the selected companies).
   • Company's stock opening price
   • Company's stock closing price 
2. Sector indices data (select the respective sector. For example if you are working on the power sector then select 'S&P BSE ENERGY'. If you cannot find the exact sector then choose the related one. For example 'S&P BSE Healthcare' for the pharmaceutical sector).
   • Sector indices closing price
   • Sector indices opening price

Methodology: After collecting the data, use the formula 'closing price minus opening price' to compute the value of stock and market return in STATA. Then perform time series regression analysis on that data with 'sectoral performance' as the independent variable and 'company performance' as the dependent variable. 

After performing the analysis you must write a 1000-word article explaining your dataset and the findings. Follow the below structure.

Article structure:

1. Introduction: Explain what is stock and market return, difference between them, and relevance of examining each return
2. Literature review: Relationship between market and stock return and how both returns are related
3. Methodology: We have a separate article on methodology. link this section to that article and explain the process of calculation in detail in this section
4. Analysis & interpretation:
   • Compuation of market and stock return along with line graph for each return and its interpretation to show how companies return is different from market return
   • Time series analysis for checking the impact of market return on stock return and computing the values of:
     • intercept 
     • slope coefficient value interepretation
5. Conclusion: What is the use of intercept and slope coefficient? here you have to identify how knowledge of these two values would enable assessment of any change by considering basic relationship between market and stock return

Your submission must include data files along with the article in a word document.

This milestone is in continuation of the milestone titled 'What are abnormal returns on stocks and how to calculate them?'

Aim: The aim of this milestone is to calculate the t-statistic and standard error values for chosen stocks which will help us test the impact of quarterly results announcement on stock peformance day-wise.

Background & Purpose: In the previous article we calculated the Expected Returns (ER), Abnormal Returns (AR) and Average Abnormal Returns (AAR). In this milestone we use those results to calculate the t-statistic for the companies (Abnormal Returns) as well as the overall sector (Average Abnormal Returns). Since we will show the effect day-wise, it will give investors a good idea about which day is the best time to buy or sell a stock when quarterly results are announced.

Data: Use the following data in this article:

• The previous article's results i.e., Abnormal Returns (AR) and Average Abnormal Returns (AAR)
• Excel sheet showing:
  • Name of the company
  • column showing Date of announcement, date of T-1 day, date of T-2 day, date of T+1 day, date of T+2 day
  • column for classification of the dates (T/ T-2, T-1, T+1, T+2)
  • AAR (there will be a single value for all cells in the column, since this is sectoral average data)
  • AR

Methodology: Import the data in STATA and calculate the following:

• average abnormal return t-test (t = 𝐴𝐴𝑅/ 𝜎(𝐴𝐴𝑅), where AAR = Average abnormal return  and σ (AAR) = Standard error of average abnormal return 
• abnormal return t-test (t = 𝐴𝑅/ 𝜎(𝐴𝑅), where AR = Abnormal return and σ (AR) = Standard error of abnormal return 

In both the above cases, standard error is calculated as SE = 𝜎 /√𝑁 

After calculating the figures, write an article explaining the results in the following structure.

Article Structure: 

• Introduction- review the previous article and explain the relevance of t-statistic in examining the linkage between result announcement and stock prices. 
• Analysis- Explain the analysis procedure undertaken to calculate the t-statistic for AR and AAR.
• Interpretation- What do the following results mean:
  • AR t-statistic
  • AAR t-statistic
  • standard error
• Conclusion- what is the overall observation and conclusion of day-wise trend/ impact of result announcement on stock prices for your specific sector?

Use graphs and figures to illustrate your point.

Use references sufficiently to support your findings. 

Submit the data files and output files along with the artile in word document.

This milestone is in continuation of the milestone titled 'Caculation and relevance of t-statistic in event study methodology'.

Aim: The aim of this milestone is to compute the descriptive statistics to examine the impact of results announcement on stock prices.

Background: In the previous milestone we calculated the t-statistic for AR (Abnormal returns) and AAR (Average Abnormal Returns) for all companies in each sector to check the result announcement effect. We also calculated the standard error (SE) to support the t-statistic value. 

Purpose: In this milestone we want to compute the descriptive statistics which is essential for performing t-test, i.e. testing our hypothesis in the final stage. Descriptive statistics give us more insight into the basic nature of the dataset.

Data: Use the following data in this milestone. This data can be obtained from the previous article.

• The previous article's results i.e., Abnormal Returns (AR) and Average Abnormal Returns (AAR)
• Excel sheet showing:
  • Name of the company
  • column showing Date of announcement, date of T-1 day, date of T-2 day, date of T+1 day, date of T+2 day
  • column for time period classification of the dates (T/ T-2, T-1, T+1, T+2)
  • AAR (there will be a single value for all cells in the column, since this is sectoral average data)
  • AR

Methodology: Use STATA to calculate the following:

• Mean
• Variance
• Standard deviation
• minimum
• maximum
• Skewness
• Kurtosis

After completing the calculations, write a 1000-1500 words article in the below structure.

• Introduction- review the previous article and explain the relevance of descriptive statistics in examining the nature of the dataset; in this case, stock market data. 
• Analysis- Explain the analysis procedure undertaken to calculate the descriptive statistics.
• Interpretation- What do the following results mean for our dataset:
  • mean
  • variance
  • standard deviation
  • minimum
  • maximum
  • skewness
  • kurtosis
• Conclusion- what is the overall observation and conclusion of day-wise trend/ nature of the data?

Use graphs and figures to illustrate your point.

Use references sufficiently to support your findings. 

Submit all data files and output files along with word document containing the article.

Purpose: This article will be a comprehensive explanation of the methodology undertaken for meeting the first part of the second goal of this study- i.e., EVENT STUDY METHODOLOGY to assess the impact of results announcement on share prices. the methodology is as follows

• step 1- data collection for energy, pharmaceutical, banking & finance sector companies listed on BSE (training data is FY 2017-2018 and test data is FY 2018-2023). explain the chosen variables too:
  • date of quarterly result announcement
  • stock opening & closing price, and high & low price for the event window i.e. quarterly result announcement date ('T'), 2 days before event (' T-1', 'T-2',) and 2 days after the event ('T+1', 'T+2').
  • stock market (BSE) opening and closing value for the event window, i.e. T-2, T-1, T, T+1, T+2.
  • Volume of trade, i.e. No. of shares traded for the company during the event window
• step 2- calculation of returns value (stock opening price minus closing price)
• step 3- regression test to calculate intercept and slope coefficient
• step 4- calculating returns for event study methodology- expected returns, average returns and abnormal average returns
• step 5- calculating t and standard error
• step 6- calculating descriptive statistics- mean, standard deviation, skewness and kurtosis 
• step 7- performing t-test
• step 8- performing paired t-test to estimate the impact 

software used- STATA (calculations) & MS Excel (data collection)

Structure: 

1. Purpose of this study- creating a prediction model for result announcement effect
2. Methodology
   • variables identified
   • time period
   • stocks/ sectors chosen
   • index chosen
   • tests applied (step wise)
   • limitations of the dataset- missing values if any, limited sectors, etc.
   • software used
   • data analysis presentation method

This milestone is in continuation of the milestone titled 'Calculating market return and its role in determining stock return'.

Aim: The aim of this milestone is to compute the abnormal return along with understanding its significance in broadening knowledge about stocks

Background: In the previous milestone we calculated the slope coefficient and intercept using time series regression. Use those results to proceed with this one. 

Purpose: The overall purpose of this goal is to predict stock prices after a company's quarterly results are announced. To understand this, we need to check whether any additional benefit could be derived or not. This additional benefit or loss is abnormal return. 

Data: Use the following data in this article:

• The previous article's results i.e., slope coefficient and intercept (1st April 2017- 31st March 2018)
• Sector indices data (select the respective sector in the link. For example if you are working on the power sector then select 'S&P BSE ENERGY'. If you cannot find the exact sector then choose the related one. For example 'S&P BSE Healthcare' for the pharmaceutical sector) for the period 1st April 2018 to 31st March 2023. 

Methodology: Using STATA, caculate the following:

• expected return
• abnormal return
• average abnormal return .

The formulae are as follows:

1. Expected Return = ER = ai + bi Rm
   • Wherein, ER = Expected return for stock at time t (take average of the 5 days wherein)
   • ai = Intercept or alpha coefficient of i th stock.
   • bi = Slope beta coefficient of i th stock.
   • Rm = Expected return on index (sectoral indices data)
2. Abnormal return= AR = R – ER
   • Wherein, ER= Expected return
   • R = Actual Returns
3. Average abnormal return= AAR = ∑ 𝐴𝑅𝑖 /𝑁 
   • Wherein, i = the number of securities in the study; N = total number of securities in the class (group)
   • This will be computed by summing all stocks abnormal return for a particular time and then dividing by no. of stocks.

Prior knowledge/ experience of using STATA is mandatory.

After completing the analysis you must write a 1000-1500 word article following the below structure.

• Introduction- explain what is abnormal return and whats its significance for making investment decision
• Analysis- Calculation of expected, abnormal, and average abnormal return
• Interepretation- of return values for different companies
• Discussion- Trend graph of abnormal and average abnormal return and its interpretation
• Conclusion- what is the status of abnormal and average abnormal return and how these values will help in deciding on announcement effect on stocks and investment decision?

Submit all the data files, output files and word document containing the article.

All conceptual explanations must be sufficiently referenced.

This milestone is in continuation of the milestone titled 'Importance of descriptive statistics in revealing stock market performance of companies'.

Aim: The aim of this milestone is to test the effect of quarterly result announcement on stock prices on daily basis.

Background: In the previous milestone we calculated the descriptive statistics (mean, variance, standard deviation, minimum, maximum, skewness, kurtosis) for the individual stocks and the overall sector. 

Purpose: In this milestone we want to test the hypothesis that quarterly results announcement has an impact on stock prices. We also want to show whether the change is increasing (i.e. rise in stock price) or decreasing (i.e. fall in stock price).

Data: Use the following data in this milestone. This data can be obtained from the previous milestone.

• Expected Returns (ER)
• Abnormal Returns (AR)
• Average Abnormal Returns (AAR)
• Abnormal Returns Test (t-statistic and standard error)
• Average Abnormal Returns Test (t-statistic and standard error)

Hypothesis: Our null hypothesis (H0): There is no impact of returns announcement on stock prices

Methodology: Simply do hypothesis testing on the previous milestone's datasheet to show the company-wise daily effect (AR) and sector-wise daily effect (AAR). For the variables having absolute t-stat value of less than 1.96, the null hypothesis won't be rejected while all absolute t-stat value of more than 1.96 the null hypothesis is rejected showing that result announcement has impact on stock price. Now looking at the abnormal return value we can see which stock has highest influence of result annpouncement. Remember that this absolute return value is checked only for significant results based companies.

After completing the analysis calculations, write a 1000-1500 words article following the below structure.

• Introduction- review the previous article and explain the t-statistic and standard error results in examining the effect of results announcement on stock price data. 
• Analysis- Explain the analysis procedure undertaken to assess the hypothesis.
• Interpretation- What do the following results mean for our dataset, i.e., how does results announcement affect daily stock prices of each company and overall sector?
  • trend analysis for significant abnormal return observations
  • trend analysis for significant average abnormal return observations
• Conclusion- what is the overall observation and conclusion of day-wise trend/ nature of the data?

Use graphs and figures to illustrate your point.

Use references sufficiently to support your findings. 

Submit all data files and output files along with the word document of your article.