Good day Jonathan. I really enjoyed watching your video. May I ask what if I need to calculate the result for 1000 observations. How can I do that since drag and drop formula for 1000 observations manually seem impossible. Many thanks if you can answer.
Hi Jonathan. Thanks for sharing. I have a queation related to expected return formula. Is there any academic article / journal which calculate expected return based on the average of actual return during a specific time. Thanks
Thanks for your video, but have a question. As I check the formula the expected return (Rit) = Rit＝ai ＋ bi Rmt (Return of the stock market for the estimated period) but your expected return was took the estimated period of stock return only?
Hi Jonathan, thanks for this help video. I just wonder whether it is fine to sum percentage returns together. If I remember correctly, only log returns can be summed.
Thank you for this great video! May i ask to be sure that 5 day car corresonding to estimation window of (0,5) and 15 day car corresponding to (-5,10) is it right?
Thanks so much, this is a brilliant video. Could you please do the statistics tests for CAAR and AAR with t and p values and statistical significance? :-)
Jonathan thank you for your video. I have applied what you have eluded in your video, however, the statistical testing i.e. deviations - how whould i know if its statistically significant or not? And can you site some literature where this technique has been used,
T Stat= Abnormal Return: Standart Error. If T stat is higher than 1,96 (and parallely lower than -1,96) it is statistically significant. As for CAR and its significance I am not Sure. Maybe the same?
Hi there, great video. I got two question. what do you use as a proxy for the expected return for bond ? how do you determine the significance test ? thank you
It is used to measure the financial performance of a firm, but what about the operational performance of a firm, can it be useful in determining the operational performance?
Hi, can could you please advise how this can be conducted for one company (or an index rather). I have multiple events I would like to tests against the DJIA
Hello Jonathan, Nice video. I have one query having calculated CAAR for the event window, how do we find whether CAAR is significant or not? There are research papers in which authors have lets say taken 11 day event window then they have 11 values of CAAR and they have calculated z/t statistics for each of these days. Can you please let me know how to find significance of each CAAR record in the event window?
Good evening. I try to calculate the -30,-1 CAR. I always find zero 0 because from the sum of the observations (returns) we abstract the sum of the mean (expected returns). That happens for any stock and any observation period even in your example. I think something is missing. Can you help?
why do you use the 5 day CAR and the 15 day? If my event study has an event window of 41 (+20 , - 20, day zero = announcement day) would i do a 20 day CAR from day 0 to day -20? Help please x
In the example, he only has data for 5 days post event which I guess is why he is using 5 day CAR, and the 15-day car is just another example I believe. The time frame you use would depend on what you want to test. Like in this example if you want to test the abnormal returns after the event you would use the 5-day but if you would like to test for example the abnormal returns leading up to the event you would have to choose a different time window.
1) Copy all dates to another sheet, and in that sheet sort and remove duplicates. Label these starting with 1. 2) You can then use a VLOOKUP on the returns table to get the initial fake date. 3) Use a VLOOKUP on the event table to get a fake date for the event. 4) Use a final VLOOKUP to match return data with the specific event and get the event time
Further steps could be done similarly. For instance, create another table for each stock average. Use SUMIF and COUNTIF functions to get an average for each stock. It takes some doing, but you can do this for a variable number of stocks
+Kamilla Islambekova It doesn't. There is only one event in this study, we are simply estimated a cumulative abnormal return around the event using a particular window.
Depends on what you want to accomplish. This exercise is to understand what an event study is and in my course this happens before understanding what expected returns are. Also, as a first order approximation it’s not terrible. Sure, I could complicate this with DGTW characteristic adjusted return with some sort of complication layered on top of it but there is no point to for the purposes of this exercise. So when you say “no, totally wrong” you’re showing your ignorance. Life advice, don’t be an a**hat to people providing a public service, especially when they’re more likely to be experts than you.
Thanks man ! If it wasnt this video, I would have been crying
thank you!.. can you please explain how to do the statistic tests for CAR and CAAR? specially how to get the standard deviation.. Thanks!
Hello, did you find a solution to your question?
@@joshbaby12ahave you found a solution to that haha? Cause I need one lol
Thanks so much Jonathan. Helped me in my project work at Uni ;)
Do you not need to test the significance of the event to the return?
Best video I've found on the topic. Thank you
Good day Jonathan. I really enjoyed watching your video. May I ask what if I need to calculate the result for 1000 observations. How can I do that since drag and drop formula for 1000 observations manually seem impossible. Many thanks if you can answer.
Hi Jonathan. Thanks for sharing. I have a queation related to expected return formula. Is there any academic article / journal which calculate expected return based on the average of actual return during a specific time. Thanks
Thanks for your video, but have a question. As I check the formula the expected return (Rit) = Rit＝ai ＋ bi Rmt (Return of the stock market for the estimated period) but your expected return was took the estimated period of stock return only?
Thanks for sharing, very easy to understand
Great video and superb explanation doctor! I have one question, how do you determine the statistical significance of the results? Thank you!
Hi Jonathan, thanks for this help video. I just wonder whether it is fine to sum percentage returns together. If I remember correctly, only log returns can be summed.
Hey Teach, why did we do an average for Expected Returns, and use that value for the quarter instead of calculating ER for each day?
Thank you for this great video! May i ask to be sure that 5 day car corresonding to estimation window of (0,5) and 15 day car corresponding to (-5,10) is it right?
I think 5 days return should be -2 to +2 and 15 days CAR should be -7 to +7
I got proper guidance by this video
I thank you so much for this video . Very helpful for my academic project .
What price or return should I take for event day if event day is a holiday for exchange?
Regarding expected return, is it possible to use the average of the actual return?
Thanks so much, this is a brilliant video. Could you please do the statistics tests for CAAR and AAR with t and p values and statistical significance? :-)
Did you find the answer?
Jonathan thank you for your video. I have applied what you have eluded in your video, however, the statistical testing i.e. deviations - how whould i know if its statistically significant or not? And can you site some literature where this technique has been used,
T Stat= Abnormal Return: Standart Error.
If T stat is higher than 1,96 (and parallely lower than -1,96) it is statistically significant.
As for CAR and its significance I am not Sure. Maybe the same?
Hi there, great video. I got two question. what do you use as a proxy for the expected return for bond ? how do you determine the significance test ? thank you
It is used to measure the financial performance of a firm, but what about the operational performance of a firm, can it be useful in determining the operational performance?
Hi, can could you please advise how this can be conducted for one company (or an index rather). I have multiple events I would like to tests against the DJIA
Hello Jonathan, Nice video. I have one query having calculated CAAR for the event window, how do we find whether CAAR is significant or not? There are research papers in which authors have lets say taken 11 day event window then they have 11 values of CAAR and they have calculated z/t statistics for each of these days. Can you please let me know how to find significance of each CAAR record in the event window?
Did you find the answer??
hi thanks for the video really helped me out, do you know if there is a way to compute standard errors for these cumulative abnormal returns, thanks
Did you find the answer,?
Good evening.
I try to calculate the -30,-1 CAR. I always find zero 0 because from the sum of the observations (returns) we abstract the sum of the mean (expected returns). That happens for any stock and any observation period even in your example. I think something is missing. Can you help?
what If i am using a 41 day window then should i leave 20 days pre and post from the event date ?? and for 30 day window 15 days?
Dear Kalodimos, do you maybe have available the excel sheet?
How can I apply this to several Stock Market Indices and several events?
why do you use the 5 day CAR and the 15 day? If my event study has an event window of 41 (+20 , - 20, day zero = announcement day) would i do a 20 day CAR from day 0 to day -20? Help please x
In the example, he only has data for 5 days post event which I guess is why he is using 5 day CAR, and the 15-day car is just another example I believe. The time frame you use would depend on what you want to test. Like in this example if you want to test the abnormal returns after the event you would use the 5-day but if you would like to test for example the abnormal returns leading up to the event you would have to choose a different time window.
what if the first event day open at 25%
If I add the AAR will it become CAAR
Very helpful, thanks a lot!
Hello, just to clarify, is AR the same as AAR? If not, how to I calculate the AAR?
Actually I'm stuck with AAR and t value
Did you find the answer
thank you very munch you helped for my idea
I have 112 events (FOMC statements) - do I still have to go through this process one by one?
1) Copy all dates to another sheet, and in that sheet sort and remove duplicates. Label these starting with 1.
2) You can then use a VLOOKUP on the returns table to get the initial fake date.
3) Use a VLOOKUP on the event table to get a fake date for the event.
4) Use a final VLOOKUP to match return data with the specific event and get the event time
Further steps could be done similarly. For instance, create another table for each stock average. Use SUMIF and COUNTIF functions to get an average for each stock. It takes some doing, but you can do this for a variable number of stocks
How would I go about applying this to government bond yields?
+Ethan Allen Same procedure. Just makes sure you are looking at changes in yields (which is sort of like a return) and not the raw yields.
how this analysis differ for the companies with one event day (catastrophy)?
+Kamilla Islambekova It doesn't. There is only one event in this study, we are simply estimated a cumulative abnormal return around the event using a particular window.
thks a lot for teaching
What if the event date is non-trading day
consider the next trading day.
How to make table?
Several mistakes. But the major one is the time. Your fake time should be overlapping the companies events
😎 😎 😎
Salamaleksus
no, totally wrong. you just take average return as expected return, that is not a way to go
Depends on what you want to accomplish. This exercise is to understand what an event study is and in my course this happens before understanding what expected returns are. Also, as a first order approximation it’s not terrible. Sure, I could complicate this with DGTW characteristic adjusted return with some sort of complication layered on top of it but there is no point to for the purposes of this exercise. So when you say “no, totally wrong” you’re showing your ignorance. Life advice, don’t be an a**hat to people providing a public service, especially when they’re more likely to be experts than you.
no need to be a tit su zhang