One of the best explanations I've found. Concise and effective! I have a question. I am analyzing indexes rather than the traditional company to a benchmark, as in your example. Specifically, I am using daily data on the G7 and BRICS Nations Indexes (CAC, FTSE, NIFTY etc) and will use the MSCI World as the 'market' benchmark/comparison. Which method would be most appropriate for this? The Market Model? Unless I am mistaken, I cannot use the RAR method given the beta and alpha is already taken care of in an index and thus would be pointless?
Can you please conduct a walkthrough for an event study of 200 M&A transactions using the Market method? How do I perform t-test on the cross-sectional average abnormal return for these 200 M&As over 200 days of estimation period and 20 days of event period?
Dear author, thank you for the video. The methodology is well described. Now I am on the stage of modelling, but in Stata, not in Excel, are you familiar with this software?
Dear Sir Wonderful explanation. Can you please share that excel sheet so that we can have a hands on approach. I tried to download the data from Yahoo but there are some discrepencies
H, i'm writting a bachelor thesis about the effects of video-game announcements on the firms share price. I wanted to ask which method you suggest for a thesis like that, or should i even do all 3 methods? Thanks & great video!
That's an interesting study! How did it turn out? I'd be interested in reading it. I'm doing a dissertation on effects of breach announcements on the firm's share price.
Hello Sir, thank you for the explanation video, i would ask something, Should the confounding event in the estimation period be excluded? or just exclude it from the event period only?
hallo!!Thnx for the tutoring is very enlightening!!Out of curiosity, couldnt we just calculate the regression in the last model and then use its residuals in order to calculate risk adjusted returns model?? thnk you very much!
Thanks! Great presentation. If the abnormal returns were positive, how would you test if they are significantly different from zero? Please make a video with this as well. Please try to organize your videos into playlists, so viewers could browse them easily.
Solomon Opare: For each company, calculate each day's AR around its own event, say 5 days before and 5 days after. Then line up the days across the firms and calculate the average AR for each day within the event window.
Hey Pat, I am trying to conduct an event study over a whole industry and across different countries including different currencies (EURO, YEN, etc.). How do I have to deal with the different currencies? Do I have to set a base unit (e.g., USD)? Thanks a lot
@@PatObi Thank you for your reply. If I have to check for abnormal returns would it be better to take previous day since our event day is holiday and then compare it with abnormal return of next trading day, that is say Tuesday.. If my original event day is Sunday(for which friday price is taken) ?
Hamdani Imed: The basic process is to divide each AR by the std error. However the std error uses parameter estimates from the estimation period. Pls confirm by taking a look at the literature. Thanks.
so for the first method, the means adjusted returns, we cannot calculate a beta, intercept and standard deviation right, because we dont compare it to a market?
@@PatObi Thanks for your answer. How do you calculate statistical significance, t and z values for the means adjusted returns model? Since I am not comparing to a market, so I dont have a beta. :S
Sir I would like to calculate a 3 day window, 10 day window and 41 day window so is there a problem if i take 41 day window for calculating the intercept, slope and then apply for all the windows the same??
The slope and intercept estimates can be used for all event windows. However be sure to calculate these estimates from the estimation period, which is a 'clean' period outside of the event windows.
You are brilliant. I wish if you make another episode but with more advanced methodology.
Who is the founder of the Market Adjusted Model? Can I please have the link to a study?
Excellent! A strong sense of detail. Thank you brother
I can’t believe I am just watching this, excellent explanation
Hi Pat, Excellent one, thanks! Are you able to share the spreadsheet as well?
Can you tell me how you calculate the Cumulative AR from the AR's? Is this just summing up the AR's? I'd like to calculate the CAR.
Excellent walkthrough!
One of the best explanations I've found. Concise and effective!
I have a question. I am analyzing indexes rather than the traditional company to a benchmark, as in your example.
Specifically, I am using daily data on the G7 and BRICS Nations Indexes (CAC, FTSE, NIFTY etc) and will use the MSCI World as the 'market' benchmark/comparison.
Which method would be most appropriate for this? The Market Model?
Unless I am mistaken, I cannot use the RAR method given the beta and alpha is already taken care of in an index and thus would be pointless?
You're on point. You can only use the mean adjusted model when using market indexes.
You can only use the mean adjusted method with indexes
Hi Joseph, I'm currently working on a similar thesis would you mind getting in touch, I'd like to address some questions.
Excellent explanation. It would be super helpful if you could provide the link to this spreadsheet.
Can you please conduct a walkthrough for an event study of 200 M&A transactions using the Market method? How do I perform t-test on the cross-sectional average abnormal return for these 200 M&As over 200 days of estimation period and 20 days of event period?
I believe I owe you my gratitude
Excellent! Please post a video on doing rest of significance of the adjusted returns.
Dear author, thank you for the video. The methodology is well described. Now I am on the stage of modelling, but in Stata, not in Excel, are you familiar with this software?
What if we were using daily data? You said you were using monthly for fun... does that mean daily is the standard?
Do you know how we can compute the daily return instead of the monsthly return ? With what value should we divide our index price?
I actually have the same question...
Dear Sir
Wonderful explanation. Can you please share that excel sheet so that we can have a hands on approach. I tried to download the data from Yahoo but there are some discrepencies
H, i'm writting a bachelor thesis about the effects of video-game announcements on the firms share price. I wanted to ask which method you suggest for a thesis like that, or should i even do all 3 methods? Thanks & great video!
For a robust study, it's better to use all methods. Risk-adjusted models are best though.
That's an interesting study! How did it turn out? I'd be interested in reading it. I'm doing a dissertation on effects of breach announcements on the firm's share price.
Great video! Really appreciated this
Dear author thank you for vedio..
Please can you make vedios with IPO event.. because IPOs have not historical prices.
Hello Sir, thank you for the explanation video, i would ask something, Should the confounding event in the estimation period be excluded?
or just exclude it from the event period only?
hallo!!Thnx for the tutoring is very enlightening!!Out of curiosity, couldnt we just calculate the regression in the last model and then use its residuals in order to calculate risk adjusted returns model?? thnk you very much!
Thank you so much! it's very useful!!
could you tell me all literatures that you used sir?
Thanks! Great presentation. If the abnormal returns were positive, how would you test if they are significantly different from zero? Please make a video with this as well. Please try to organize your videos into playlists, so viewers could browse them easily.
Great video, this has really helped me with my research :)
How do you calculate the three different returns when you have events from more than one company?
Solomon Opare: For each company, calculate each day's AR around its own event, say 5 days before and 5 days after. Then line up the days across the firms and calculate the average AR for each day within the event window.
thanks
Hey Pat, I am trying to conduct an event study over a whole industry and across different countries including different currencies (EURO, YEN, etc.). How do I have to deal with the different currencies? Do I have to set a base unit (e.g., USD)?
Thanks a lot
That's what I'd do - using the coincident FX for each period.
the standard deviation pls, how did you calculate it?
What price or return should I take for event day if event day is a holiday for exchange?
Up to you. But generally, the next trading day.
@@PatObi Thank you for your reply. If I have to check for abnormal returns would it be better to take previous day since our event day is holiday and then compare it with abnormal return of next trading day, that is say Tuesday.. If my original event day is Sunday(for which friday price is taken) ?
Thanks!
Hello, thanks for the video. I would like to know how do you for testing the significance of AR on each day in the event window ?
Hamdani Imed: The basic process is to divide each AR by the std error. However the std error uses parameter estimates from the estimation period. Pls confirm by taking a look at the literature. Thanks.
hey pat, can i use the simple return (G5/G4)-1 instead of the continuous compounding ln(G5/G4) ???
For empirical analysis, the log form is more appropriate because it has better distributional properties.
how and where do you get S&P or any market returns for that time?
There are several databases from which you can obtain index data.
so for the first method, the means adjusted returns, we cannot calculate a beta, intercept and standard deviation right, because we dont compare it to a market?
Yes.
@@PatObi Thanks for your answer. How do you calculate statistical significance, t and z values for the means adjusted returns model?
Since I am not comparing to a market, so I dont have a beta. :S
Sir I would like to calculate a 3 day window, 10 day window and 41 day window so is there a problem if i take 41 day window for calculating the intercept, slope and then apply for all the windows the same??
The slope and intercept estimates can be used for all event windows. However be sure to calculate these estimates from the estimation period, which is a 'clean' period outside of the event windows.
can you share excel sheet ......
excel bhej bhai